diff --git a/CURRENT_STATE.md b/CURRENT_STATE.md
index d91b1fe..77c9678 100644
--- a/CURRENT_STATE.md
+++ b/CURRENT_STATE.md
@@ -1,6 +1,6 @@
 # H3 Runtime Current State
 
-Status date: 2026-08-24
+Status date: 2026-08-25
 
 This document is the canonical snapshot of implemented scope and remaining work.
 Historical handoffs in `PLAN.md` and `PARITY.md` may describe older states.
@@ -160,6 +160,18 @@ video and audio tensors. Spark enables the path with
 `SAGE2_BLACKWELL_DESIGN.md`, and
 `benchmarks/gb10-post-optimization-profile-summary.json`.
 
+The isolated FC2 cuBLASLt scheduling study is complete. The production
+heuristic's `_stream_k` kernel requests the same `25.664 GB` of operands as the
+retained public split-K-1 schedule, but its L2 hit rate is only `53.32%` versus
+`91.10%`; it incurs `9.853 GB` more L2 read misses and spends heavily in
+synchronization polling. Algorithm 70, tile 20, stages 37, split-K 1 is
+byte-exact with zero workspace. It improves complete blocks 0, 24, and 49 by
+`8.16-8.88%`, the two-step trajectory by `7.50%`, and the canonical 12-step
+trajectory from `278.201 s` to `255.371 s` (`8.21%`) with exact video and audio
+latents. This remains a research-retained integration candidate: production
+dispatch and configuration are unchanged. See
+`research/fc2_nvfp4_scheduling/RESULTS.md`.
+
 The Spark hot runtime was rebuilt and recreated with image
 `sha256:5f879c43374bcedb89745971d7d95d82afc8fcf9c41d30f11b257c2863b9fe28`.
 Health and startup warmup pass with modulation fusion enabled. A resident real
diff --git a/PERFORMANCE_ROADMAP.md b/PERFORMANCE_ROADMAP.md
index 9390832..f29de8c 100644
--- a/PERFORMANCE_ROADMAP.md
+++ b/PERFORMANCE_ROADMAP.md
@@ -200,7 +200,8 @@ another's result.
 
 ### Phase 3: NVFP4 Fused Projection Prototype
 
-Status: active next component on GB10/SM121.
+Status: active on GB10/SM121; isolated FC2 library scheduling is complete and
+awaits production integration.
 
 The first profile also exposed and fixed a native activation-packer layout bug:
 the previous width-specific block-scale swizzle failed at the attention output's
@@ -383,6 +384,24 @@ off-chip traffic bottleneck, the NVFP4 GEMMs. Do not select a new kernel from
 the old profile. See
 `benchmarks/gb10-fully-fused-fresh-nsight-summary.json`.
 
+The FC2 follow-up resolves that projection's `16.1x` off-chip amplification.
+The production heuristic launches an undocumented-sentinel `_stream_k` kernel;
+the retained documented configuration is cuBLASLt algorithm 70, tile 20,
+stages 37, public split-K 1, reduction scheme 0, and zero workspace. Both
+schedules request the same `25.664 GB` of operands, but split-K 1 raises L2 hit
+rate from `53.32%` to `91.10%`, removes `9.853 GB` of L2 read misses, and raises
+tensor-pipe activity from `25.01%` to `82.42%`. There is no material global
+partial-accumulator or output-reduction traffic; the baseline instead spends
+heavily in Stream-K synchronization polling and loses traversal locality.
+
+The candidate is byte-exact and improves FC2 p50 from `52.521 ms` to
+`15.636 ms`. Complete blocks 0, 24, and 49 improve by `8.16-8.88%`; two-step
+and canonical 12-step trajectories improve by `7.50%` and `8.21%`, with exact
+video and audio latents. The custom persistent-kernel branch is therefore
+closed. Production integration remains separate work, so current dispatch is
+unchanged. See `research/fc2_nvfp4_scheduling/RESULTS.md` and
+`benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json`.
+
 The follow-up real block-24 Sage2 decomposition now selects the next exact
 kernel experiment. Manual preparation plus the existing prequantized mainloop
 is byte-exact against public SageAttention 2.2.0. Uninstrumented median phase
diff --git a/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv b/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv
new file mode 100644
index 0000000..1b17848
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv
@@ -0,0 +1,4 @@
+"ID","Process ID","Process Name","Host Name","Kernel Name","Context","Stream","Block Size","Grid Size","Device","CC","c2clink__enabled_mask","c2clink__present","derived__avg_thread_executed","derived__avg_thread_executed_true","derived__avg_thread_unexecuted_true","derived__derivative_avg_thread_executed_true","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.peak_sustained","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.per_second","derived__local_spilling_requests","derived__local_spilling_requests_pct","derived__lts__lts2xbar_bytes.sum.peak_sustained","derived__lts__lts2xbar_bytes.sum.per_second","derived__memory_l1_conflicts_shared_nway","derived__memory_l1_wavefronts_shared_excessive","derived__memory_l2_theoretical_sectors_global_excessive","derived__pct_occupancy_per_barrier_count","derived__pct_occupancy_per_block_size","derived__pct_occupancy_per_register_count","derived__pct_occupancy_per_shared_mem_size","derived__sm__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__inst_executed_op_branch_pct","derived__smsp__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hadd_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__sass_thread_inst_executed_op_hmul_pred_on_x2","derived_tempMetric0","derived_tempMetric1","derived_tempMetric2","derived_tempMetric3","derived_tempMetric4","derived_tempMetric5","device__attribute_architecture","device__attribute_async_engine_count","device__attribute_can_flush_remote_writes","device__attribute_can_map_host_memory","device__attribute_can_tex2d_gather","device__attribute_can_use_64_bit_stream_mem_ops","device__attribute_can_use_64_bit_stream_mem_ops_v1","device__attribute_can_use_host_pointer_for_registered_mem","device__attribute_can_use_stream_mem_ops_v1","device__attribute_can_use_stream_wait_value_nor","device__attribute_can_use_stream_wait_value_nor_v1","device__attribute_chip","device__attribute_clock_rate","device__attribute_cluster_launch","device__attribute_compute_capability_major","device__attribute_compute_capability_minor","device__attribute_compute_mode","device__attribute_compute_preemption_supported","device__attribute_concurrent_kernels","device__attribute_concurrent_managed_access","device__attribute_confidential_computing_mode","device__attribute_cooperative_launch","device__attribute_cooperative_multi_device_launch","device__attribute_deferred_mapping_cuda_array_supported","device__attribute_device_index","device__attribute_direct_managed_mem_access_from_host","device__attribute_display_name","device__attribute_dma_buf_supported","device__attribute_ecc_enabled","device__attribute_fb_bus_width","device__attribute_fbp_count","device__attribute_generic_compression_supported","device__attribute_global_l1_cache_supported","device__attribute_global_memory_bus_width","device__attribute_gpu_direct_rdma_flush_writes_options","device__attribute_gpu_direct_rdma_supported","device__attribute_gpu_direct_rdma_with_cuda_vmm_supported","device__attribute_gpu_direct_rdma_writes_ordering","device__attribute_gpu_overlap","device__attribute_gpu_pci_device_id","device__attribute_gpu_pci_ext_device_id","device__attribute_gpu_pci_ext_downstream_link_rate","device__attribute_gpu_pci_ext_downstream_link_width","device__attribute_gpu_pci_ext_gen","device__attribute_gpu_pci_ext_gpu_gen","device__attribute_gpu_pci_ext_gpu_link_rate","device__attribute_gpu_pci_ext_gpu_link_width","device__attribute_gpu_pci_revision_id","device__attribute_gpu_pci_sub_system_id","device__attribute_handle_type_fabric_supported","device__attribute_handle_type_posix_file_descriptor_supported","device__attribute_handle_type_win32_handle_supported","device__attribute_handle_type_win32_kmt_handle_supported","device__attribute_host_native_atomic_supported","device__attribute_host_numa_id","device__attribute_host_register_supported","device__attribute_implementation","device__attribute_integrated","device__attribute_ipc_event_supported","device__attribute_kernel_exec_timeout","device__attribute_l2_cache_size","device__attribute_l2s_count","device__attribute_limits_max_cta_per_sm","device__attribute_limits_num_tpcs","device__attribute_local_l1_cache_supported","device__attribute_managed_memory","device__attribute_max_access_policy_window_size","device__attribute_max_block_dim_x","device__attribute_max_block_dim_y","device__attribute_max_block_dim_z","device__attribute_max_blocks_per_multiprocessor","device__attribute_max_gpu_frequency_khz","device__attribute_max_grid_dim_x","device__attribute_max_grid_dim_y","device__attribute_max_grid_dim_z","device__attribute_max_ipc_per_multiprocessor","device__attribute_max_ipc_per_scheduler","device__attribute_max_mem_frequency_khz","device__attribute_max_persisting_l2_cache_size","device__attribute_max_pitch","device__attribute_max_registers_per_block","device__attribute_max_registers_per_multiprocessor","device__attribute_max_registers_per_thread","device__attribute_max_shared_memory_per_block","device__attribute_max_shared_memory_per_block_optin","device__attribute_max_shared_memory_per_multiprocessor","device__attribute_max_threads_per_block","device__attribute_max_threads_per_multiprocessor","device__attribute_max_warps_per_multiprocessor","device__attribute_max_warps_per_scheduler","device__attribute_maximum_surface1d_layered_layers","device__attribute_maximum_surface1d_layered_width","device__attribute_maximum_surface1d_width","device__attribute_maximum_surface2d_height","device__attribute_maximum_surface2d_layered_height","device__attribute_maximum_surface2d_layered_layers","device__attribute_maximum_surface2d_layered_width","device__attribute_maximum_surface2d_width","device__attribute_maximum_surface3d_depth","device__attribute_maximum_surface3d_height","device__attribute_maximum_surface3d_width","device__attribute_maximum_surfacecubemap_layered_layers","device__attribute_maximum_surfacecubemap_layered_width","device__attribute_maximum_surfacecubemap_width","device__attribute_maximum_texture1d_layered_layers","device__attribute_maximum_texture1d_layered_width","device__attribute_maximum_texture1d_linear_width","device__attribute_maximum_texture1d_mipmapped_width","device__attribute_maximum_texture1d_width","device__attribute_maximum_texture2d_gather_height","device__attribute_maximum_texture2d_gather_width","device__attribute_maximum_texture2d_height","device__attribute_maximum_texture2d_layered_height","device__attribute_maximum_texture2d_layered_layers","device__attribute_maximum_texture2d_layered_width","device__attribute_maximum_texture2d_linear_height","device__attribute_maximum_texture2d_linear_pitch","device__attribute_maximum_texture2d_linear_width","device__attribute_maximum_texture2d_mipmapped_height","device__attribute_maximum_texture2d_mipmapped_width","device__attribute_maximum_texture2d_width","device__attribute_maximum_texture3d_depth","device__attribute_maximum_texture3d_depth_alternate","device__attribute_maximum_texture3d_height","device__attribute_maximum_texture3d_height_alternate","device__attribute_maximum_texture3d_width","device__attribute_maximum_texture3d_width_alternate","device__attribute_maximum_texturecubemap_layered_layers","device__attribute_maximum_texturecubemap_layered_width","device__attribute_maximum_texturecubemap_width","device__attribute_mem_sync_domain_count","device__attribute_memory_clock_rate","device__attribute_memory_pools_supported","device__attribute_mempool_supported_handle_types","device__attribute_mps_enabled","device__attribute_multi_gpu_board","device__attribute_multi_gpu_board_group_id","device__attribute_multicast_supported","device__attribute_multiprocessor_count","device__attribute_num_l2s_per_fbp","device__attribute_num_schedulers_per_multiprocessor","device__attribute_num_tex_per_multiprocessor","device__attribute_numa_config","device__attribute_pageable_memory_access","device__attribute_pageable_memory_access_uses_host_page_tables","device__attribute_pci_bus_id","device__attribute_pci_device_id","device__attribute_pci_domain_id","device__attribute_ram_location","device__attribute_ram_type","device__attribute_reserved_shared_memory_per_block","device__attribute_sass_level","device__attribute_single_to_double_precision_perf_ratio","device__attribute_sparse_cuda_array_supported","device__attribute_stream_priorities_supported","device__attribute_surface_alignment","device__attribute_tcc_driver","device__attribute_tensor_map_access_supported","device__attribute_texture_alignment","device__attribute_texture_pitch_alignment","device__attribute_total_constant_memory","device__attribute_total_memory","device__attribute_unified_addressing","device__attribute_unified_function_pointers","device__attribute_virtual_address_management_supported","device__attribute_warp_size","gcc__cache_requests_type_constant.sum","gcc__cache_requests_type_constant.sum.pct_of_peak_sustained_elapsed","gcc__cache_requests_type_instruction.sum","gcc__cache_requests_type_instruction.sum.pct_of_peak_sustained_elapsed","gcc__xbar2gcc_sectors.sum","gcc__xbar2gcc_sectors.sum.pct_of_peak_sustained_elapsed","gpc__cycles_elapsed.avg","gpc__cycles_elapsed.avg.per_second","gpc__cycles_elapsed.max","gpc__cycles_elapsed.max.per_second","gpc__cycles_elapsed.min","gpc__cycles_elapsed.min.per_second","gpc__cycles_elapsed.sum","gpc__cycles_elapsed.sum.per_second","gpu__compute_memory_access_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.sum.pct_of_peak_sustained_elapsed","gpu__time_duration.avg","gpu__time_duration.max","gpu__time_duration.min","gpu__time_duration.sum","gr__workids_granted.avg","gr__workids_granted.max","gr__workids_granted.min","gr__workids_granted.sum","gr__workids_granted_as_ctas.avg","gr__workids_granted_as_ctas.max","gr__workids_granted_as_ctas.min","gr__workids_granted_as_ctas.sum","gr__workids_requested.avg","gr__workids_requested.max","gr__workids_requested.min","gr__workids_requested.sum","idc__request_cycles_active.avg.pct_of_peak_sustained_elapsed","idc__request_cycles_active.max.pct_of_peak_sustained_elapsed","idc__request_cycles_active.min.pct_of_peak_sustained_elapsed","idc__request_cycles_active.sum.pct_of_peak_sustained_elapsed","idc__request_hit_rate.pct","idc__requests.sum","idc__requests.sum.pct_of_peak_sustained_elapsed","inst_executed","l1tex__cycles_active.avg","l1tex__cycles_active.max","l1tex__cycles_active.min","l1tex__cycles_active.sum","l1tex__cycles_elapsed.avg","l1tex__cycles_elapsed.avg.per_second","l1tex__cycles_elapsed.max","l1tex__cycles_elapsed.max.per_second","l1tex__cycles_elapsed.min","l1tex__cycles_elapsed.min.per_second","l1tex__cycles_elapsed.sum","l1tex__cycles_elapsed.sum.per_second","l1tex__data_bank_conflicts_pipe_lsu_mem_shared.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_atom.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ld.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ldgsts.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_st.sum","l1tex__data_bank_reads.avg.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.max.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.min.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.sum.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.avg.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.max.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.min.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_atom.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.avg.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.max.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.min.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active_mem_lgds.sum","l1tex__lsu_writeback_active_mem_lgds.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active_mem_lgds.sum.peak_sustained","l1tex__lsu_writeback_active_mem_lgds.sum.per_second","l1tex__lsuin_requests.avg.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.max.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.min.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.avg.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.max.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.min.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes.sum","l1tex__m_l1tex2xbar_write_bytes.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_lg_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_lg_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes.sum","l1tex__m_xbar2l1tex_read_bytes.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes.sum.per_second","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum.per_second","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors.avg.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.max.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.min.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_red.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_st.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_lg_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_lg_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_texture.sum","l1tex__m_xbar2l1tex_read_sectors_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.per_second","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_atom.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_red.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_atom.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_ld.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_red.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_st.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_texture.sum","l1tex__t_output_wavefronts_pipe_tex_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum","l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_global_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_red.sum","l1tex__t_requests_pipe_lsu_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_st.sum","l1tex__t_requests_pipe_lsu_mem_global_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_local_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_local_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_local_op_st.sum","l1tex__t_requests_pipe_lsu_mem_local_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_atom.sum","l1tex__t_requests_pipe_tex_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_ld.sum","l1tex__t_requests_pipe_tex_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_red.sum","l1tex__t_requests_pipe_tex_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_st.sum","l1tex__t_requests_pipe_tex_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_texture.sum","l1tex__t_requests_pipe_tex_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_sector_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_atom_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_ld_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_red_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_st_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_local_op_ld_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_local_op_st_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_atom_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_ld_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_red_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_st_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_texture_op_tex_hit_rate.pct","l1tex__t_sectors_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_redas_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_texture.sum","l1tex__t_sectors_pipe_tex_mem_texture_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_texture_lookup_miss.sum","l1tex__tex_writeback_active.avg.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.max.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.min.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.sum","l1tex__tex_writeback_active.sum.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.avg.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.max.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.min.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.sum.pct_of_peak_sustained_elapsed","l1tex__throughput.avg.pct_of_peak_sustained_active","l1tex__throughput.avg.pct_of_peak_sustained_elapsed","l1tex__throughput.max.pct_of_peak_sustained_active","l1tex__throughput.max.pct_of_peak_sustained_elapsed","l1tex__throughput.min.pct_of_peak_sustained_active","l1tex__throughput.min.pct_of_peak_sustained_elapsed","l1tex__throughput.sum.pct_of_peak_sustained_active","l1tex__throughput.sum.pct_of_peak_sustained_elapsed","launch__barrier_count","launch__block_dim_x","launch__block_dim_y","launch__block_dim_z","launch__block_size","launch__cluster_dim_x","launch__cluster_dim_y","launch__cluster_dim_z","launch__cluster_max_active","launch__cluster_max_potential_size","launch__cluster_scheduling_policy","launch__cluster_size","launch__context_id","launch__device_id","launch__func_cache_config","launch__function_pcs","launch__grid_dim_x","launch__grid_dim_y","launch__grid_dim_z","launch__grid_size","launch__kernel_name","launch__occupancy_cluster_gpu_pct","launch__occupancy_cluster_pct","launch__occupancy_limit_barriers","launch__occupancy_limit_blocks","launch__occupancy_limit_registers","launch__occupancy_limit_shared_mem","launch__occupancy_limit_warps","launch__occupancy_per_barrier_count","launch__occupancy_per_block_size","launch__occupancy_per_cluster_size","launch__occupancy_per_register_count","launch__occupancy_per_shared_mem_size","launch__persisting_l2_cache_size","launch__preferred_cluster_dim_x","launch__preferred_cluster_dim_y","launch__preferred_cluster_dim_z","launch__preferred_cluster_size","launch__registers_per_thread","launch__registers_per_thread_allocated","launch__shared_mem_config_size","launch__shared_mem_per_block","launch__shared_mem_per_block_allocated","launch__shared_mem_per_block_driver","launch__shared_mem_per_block_dynamic","launch__shared_mem_per_block_static","launch__sm_count","launch__stack_size","launch__stream_id","launch__thread_count","launch__tpc_count","launch__tpc_enabled","launch__uses_cdp","launch__uses_green_context","launch__uses_mps","launch__uses_nvlink_centric_scheduling","launch__uses_vgpu","launch__waves_per_multiprocessor","lrc__average_ilc_input_sector_success_rate.pct","lrc__ilc_input_sectors.sum","lts__average_gcomp_input_sector_success_rate.pct","lts__average_gcomp_output_sector_compression_achieved_rate.ratio","lts__average_t_sectors_requested_srcunit_tex_op_read_vs_returned_sectors_realtime.ratio","lts__average_xcomp_gxc_sector_input_compression_rate.ratio","lts__cycles_active.avg","lts__cycles_active.max","lts__cycles_active.min","lts__cycles_active.sum","lts__cycles_elapsed.avg","lts__cycles_elapsed.avg.per_second","lts__cycles_elapsed.max","lts__cycles_elapsed.max.per_second","lts__cycles_elapsed.min","lts__cycles_elapsed.min.per_second","lts__cycles_elapsed.sum","lts__cycles_elapsed.sum.per_second","lts__d_atomic_input_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.max.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.min.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.avg.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.max.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.min.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.sum.pct_of_peak_sustained_elapsed","lts__d_sectors.avg.pct_of_peak_sustained_elapsed","lts__d_sectors.max.pct_of_peak_sustained_elapsed","lts__d_sectors.min.pct_of_peak_sustained_elapsed","lts__d_sectors.sum.pct_of_peak_sustained_elapsed","lts__d_sectors_fill_sysmem.sum","lts__d_sectors_fill_sysmem.sum.pct_of_peak_sustained_elapsed","lts__d_sectors_fill_sysmem.sum.per_second","lts__gcomp_input_sectors.avg","lts__gcomp_input_sectors.max","lts__gcomp_input_sectors.min","lts__gcomp_input_sectors.sum","lts__lts2xbar_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.avg.peak_sustained","lts__lts2xbar_cycles_active.avg.per_second","lts__lts2xbar_cycles_active.max.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.max.peak_sustained","lts__lts2xbar_cycles_active.max.per_second","lts__lts2xbar_cycles_active.min.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.min.peak_sustained","lts__lts2xbar_cycles_active.min.per_second","lts__lts2xbar_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.sum.peak_sustained","lts__lts2xbar_cycles_active.sum.per_second","lts__t_requests.sum","lts__t_requests_srcunit_gcc.sum","lts__t_requests_srcunit_tex.sum","lts__t_requests_srcunit_tex_op_atom_dot_alu.sum","lts__t_requests_srcunit_tex_op_atom_dot_cas.sum","lts__t_requests_srcunit_tex_op_read.sum","lts__t_requests_srcunit_tex_op_red.sum","lts__t_requests_srcunit_tex_op_write.sum","lts__t_sector_hit_rate.pct","lts__t_sector_op_read_hit_rate.pct","lts__t_sector_op_write_hit_rate.pct","lts__t_sectors.avg","lts__t_sectors.avg.pct_of_peak_sustained_elapsed","lts__t_sectors.avg.peak_sustained","lts__t_sectors.avg.per_cycle_elapsed","lts__t_sectors.max","lts__t_sectors.max.pct_of_peak_sustained_elapsed","lts__t_sectors.min","lts__t_sectors.min.pct_of_peak_sustained_elapsed","lts__t_sectors.sum","lts__t_sectors.sum.pct_of_peak_sustained_elapsed","lts__t_sectors.sum.per_second","lts__t_sectors_aperture_sysmem_op_write.sum","lts__t_sectors_aperture_sysmem_op_write.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_aperture_sysmem_op_write.sum.per_second","lts__t_sectors_data_ecc.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_data_ecc.avg.peak_sustained","lts__t_sectors_data_ecc.avg.per_cycle_elapsed","lts__t_sectors_data_ecc.sum","lts__t_sectors_data_ecc.sum.per_second","lts__t_sectors_evict_first_lookup_hit.sum","lts__t_sectors_evict_first_lookup_miss.sum","lts__t_sectors_evict_last_lookup_hit.sum","lts__t_sectors_evict_last_lookup_miss.sum","lts__t_sectors_evict_normal_demote_lookup_hit.sum","lts__t_sectors_evict_normal_demote_lookup_miss.sum","lts__t_sectors_evict_normal_lookup_hit.sum","lts__t_sectors_evict_normal_lookup_miss.sum","lts__t_sectors_lookup_hit.sum","lts__t_sectors_lookup_miss.sum","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum.per_second","lts__t_sectors_srcunit_gcc.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_gcc.avg.peak_sustained","lts__t_sectors_srcunit_gcc.avg.per_cycle_elapsed","lts__t_sectors_srcunit_gcc.sum","lts__t_sectors_srcunit_gcc.sum.per_second","lts__t_sectors_srcunit_gcc_lookup_hit.sum","lts__t_sectors_srcunit_gcc_lookup_miss.sum","lts__t_sectors_srcunit_tex.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.avg.peak_sustained","lts__t_sectors_srcunit_tex.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex.max.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.min.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.sum","lts__t_sectors_srcunit_tex.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.sum.per_second","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.avg","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.max","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.min","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.sum","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.avg","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.max","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.min","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_lookup_hit.sum","lts__t_sectors_srcunit_tex_lookup_miss.avg","lts__t_sectors_srcunit_tex_lookup_miss.max","lts__t_sectors_srcunit_tex_lookup_miss.min","lts__t_sectors_srcunit_tex_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom.sum","lts__t_sectors_srcunit_tex_op_atom.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_alu.sum","lts__t_sectors_srcunit_tex_op_atom_dot_alu.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_alu_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_dot_alu_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_cas.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_cas_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_read.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_read.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_read.sum","lts__t_sectors_srcunit_tex_op_read.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_read.sum.per_second","lts__t_sectors_srcunit_tex_op_read_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_red.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_red.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_red.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_red.sum","lts__t_sectors_srcunit_tex_op_red.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_red.sum.per_second","lts__t_sectors_srcunit_tex_op_red_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_red_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_write.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_write.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_write.sum","lts__t_sectors_srcunit_tex_op_write.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_write.sum.per_second","lts__t_sectors_srcunit_tex_op_write_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_lookup_miss.sum","lts__t_tag_requests.avg.pct_of_peak_sustained_elapsed","lts__t_tag_requests.max.pct_of_peak_sustained_elapsed","lts__t_tag_requests.min.pct_of_peak_sustained_elapsed","lts__t_tag_requests.sum.pct_of_peak_sustained_elapsed","lts__throughput.avg.pct_of_peak_sustained_elapsed","lts__throughput.max.pct_of_peak_sustained_elapsed","lts__throughput.min.pct_of_peak_sustained_elapsed","lts__throughput.sum.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.max.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.min.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_input.sum","lts__xcomp_gxc_sectors_input.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_input.sum.per_second","lts__xcomp_gxc_sectors_output.sum","lts__xcomp_gxc_sectors_output.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_output.sum.per_second","memory_access_size_type","memory_access_type","memory_l1_tag_requests_global","memory_l1_wavefronts_shared","memory_l1_wavefronts_shared_ideal","memory_l2_theoretical_sectors_global","memory_l2_theoretical_sectors_global_ideal","memory_l2_theoretical_sectors_local","memory_type","numa__cpu_affinity","numa__dev_display_name_all","numa__id_cpu","numa__id_memory","nvlink__bandwidth","nvlink__count_logical","nvlink__count_physical","nvlink__destination_ports","nvlink__dev0Id","nvlink__dev0type","nvlink__dev1Id","nvlink__dev1type","nvlink__dev_display_name_all","nvlink__enabled_mask","nvlink__is_direct_link","nvlink__is_nvswitch_connected","nvlink__max_count","nvlink__peer_access","nvlink__peer_atomic","nvlink__source_ports","nvlink__system_access","nvlink__system_atomic","pmsampling:gr__ctas_launched_queue_sync_realtime.sum","pmsampling:l1tex__data_pipe_lsu_wavefronts.avg","pmsampling:l1tex__lsu_writeback_active.avg","pmsampling:l1tex__t_sector_hit_rate.pct","pmsampling:lts__average_t_sector_hit_rate_realtime.pct","pmsampling:sm__average_thread_inst_executed_pred_on_per_inst_executed_realtime.pct","pmsampling:sm__cycles_active.avg","pmsampling:sm__inst_executed_pipe_alu_realtime.avg.pct_of_peak_sustained_elapsed","pmsampling:sm__inst_executed_realtime.avg.per_cycle_active","pmsampling:sm__pipe_tensor_cycles_active_realtime.avg.pct_of_peak_sustained_elapsed","pmsampling:smsp__inst_executed_pipe_fmaheavy.avg.pct_of_peak_sustained_elapsed","pmsampling:smsp__inst_executed_pipe_fmalite.avg.pct_of_peak_sustained_elapsed","pmsampling:tpc__warps_active_realtime.avg.per_cycle_active","pmsampling:tpc__warps_active_realtime.sum.per_cycle_active","profiler__perfworks_session_reuse","profiler__pmsampler_buffer_size_bytes","profiler__pmsampler_ctxsw_0","profiler__pmsampler_ctxsw_1","profiler__pmsampler_interval_time","profiler__pmsampler_merged_samples","profiler__pmsampler_pass_groups","profiler__replayer_bytes_mem_accessible.avg","profiler__replayer_bytes_mem_accessible.max","profiler__replayer_bytes_mem_accessible.min","profiler__replayer_bytes_mem_accessible.sum","profiler__replayer_bytes_mem_backed_up.avg","profiler__replayer_bytes_mem_backed_up.max","profiler__replayer_bytes_mem_backed_up.min","profiler__replayer_bytes_mem_backed_up.sum","profiler__replayer_passes","profiler__replayer_passes_type_warmup","profiler__timestamp_workload_end_0","profiler__timestamp_workload_end_1","profiler__timestamp_workload_start_0","profiler__timestamp_workload_start_1","sass__inst_executed_global_loads","sass__inst_executed_global_stores","sass__inst_executed_local_loads","sass__inst_executed_local_stores","sass__inst_executed_per_opcode","sass__inst_executed_per_opcode_category","sass__inst_executed_per_opcode_with_modifier_all","sass__inst_executed_per_opcode_with_modifier_selective","sass__inst_executed_register_spilling","sass__inst_executed_register_spilling_mem_local","sass__inst_executed_register_spilling_mem_shared","sass__inst_executed_register_spilling_op_read","sass__inst_executed_register_spilling_op_write","sass__inst_executed_shared_loads","sass__inst_executed_shared_stores","sass__thread_inst_executed_per_opcode_category","sass__thread_inst_executed_true_per_opcode","sass__thread_inst_executed_true_per_opcode_with_modifier_all","sass__thread_inst_executed_true_per_opcode_with_modifier_selective","sm__cycles_active.avg","sm__cycles_active.max","sm__cycles_active.min","sm__cycles_active.sum","sm__cycles_elapsed.avg","sm__cycles_elapsed.avg.per_second","sm__cycles_elapsed.max","sm__cycles_elapsed.max.per_second","sm__cycles_elapsed.min","sm__cycles_elapsed.min.per_second","sm__cycles_elapsed.sum","sm__cycles_elapsed.sum.per_second","sm__dcc_request_hit_rate.pct","sm__dcc_requests.sum","sm__dcc_requests.sum.pct_of_peak_sustained_elapsed","sm__dcc_requests_lookup_hit.sum","sm__icc_request_hit_rate.pct","sm__icc_requests.sum","sm__icc_requests.sum.pct_of_peak_sustained_elapsed","sm__inst_executed.avg.pct_of_peak_sustained_elapsed","sm__inst_executed.avg.per_cycle_active","sm__inst_executed.avg.per_cycle_elapsed","sm__inst_executed.max.pct_of_peak_sustained_elapsed","sm__inst_executed.max.per_cycle_active","sm__inst_executed.max.per_cycle_elapsed","sm__inst_executed.min.pct_of_peak_sustained_elapsed","sm__inst_executed.min.per_cycle_active","sm__inst_executed.min.per_cycle_elapsed","sm__inst_executed.sum.pct_of_peak_sustained_elapsed","sm__inst_executed.sum.per_cycle_active","sm__inst_executed.sum.per_cycle_elapsed","sm__inst_executed_pipe_adu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma_type_fp16.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma_type_fp16.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tensor_subpipe_hmma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tensor_subpipe_hmma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tensor_subpipe_imma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tensor_subpipe_imma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.sum.pct_of_peak_sustained_elapsed","sm__inst_issued.avg.pct_of_peak_sustained_elapsed","sm__inst_issued.avg.per_cycle_active","sm__inst_issued.max.pct_of_peak_sustained_elapsed","sm__inst_issued.max.per_cycle_active","sm__inst_issued.min.pct_of_peak_sustained_elapsed","sm__inst_issued.min.per_cycle_active","sm__inst_issued.sum.pct_of_peak_sustained_elapsed","sm__inst_issued.sum.per_cycle_active","sm__instruction_throughput.avg.pct_of_peak_sustained_elapsed","sm__instruction_throughput.max.pct_of_peak_sustained_elapsed","sm__instruction_throughput.min.pct_of_peak_sustained_elapsed","sm__instruction_throughput.sum.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","sm__issue_active.avg.pct_of_peak_sustained_elapsed","sm__issue_active.max.pct_of_peak_sustained_elapsed","sm__issue_active.min.pct_of_peak_sustained_elapsed","sm__issue_active.sum.pct_of_peak_sustained_elapsed","sm__maximum_warps_avg_per_active_cycle","sm__maximum_warps_per_active_cycle_pct","sm__memory_throughput.avg.pct_of_peak_sustained_elapsed","sm__memory_throughput.max.pct_of_peak_sustained_elapsed","sm__memory_throughput.min.pct_of_peak_sustained_elapsed","sm__memory_throughput.sum.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.avg.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.max.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.min.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.sum.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.avg.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.max.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.min.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.sum.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.max.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.min.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.max.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.min.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.avg","sm__ops_path_tensor_src_bf16_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.avg.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.max","sm__ops_path_tensor_src_bf16_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.max.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.min","sm__ops_path_tensor_src_bf16_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.min.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.sum","sm__ops_path_tensor_src_bf16_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.sum.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.avg","sm__ops_path_tensor_src_fp16_dst_fp16.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.avg.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.avg.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.max","sm__ops_path_tensor_src_fp16_dst_fp16.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.max.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.max.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.min","sm__ops_path_tensor_src_fp16_dst_fp16.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.min.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.min.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.sum","sm__ops_path_tensor_src_fp16_dst_fp16.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.sum.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.sum.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.avg","sm__ops_path_tensor_src_fp16_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.avg.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.max","sm__ops_path_tensor_src_fp16_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.max.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.min","sm__ops_path_tensor_src_fp16_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.min.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.sum","sm__ops_path_tensor_src_fp16_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.sum.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp64.avg","sm__ops_path_tensor_src_fp64.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.avg.peak_sustained","sm__ops_path_tensor_src_fp64.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.avg.per_second","sm__ops_path_tensor_src_fp64.max","sm__ops_path_tensor_src_fp64.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.max.peak_sustained","sm__ops_path_tensor_src_fp64.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.max.per_second","sm__ops_path_tensor_src_fp64.min","sm__ops_path_tensor_src_fp64.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.min.peak_sustained","sm__ops_path_tensor_src_fp64.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.min.per_second","sm__ops_path_tensor_src_fp64.sum","sm__ops_path_tensor_src_fp64.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.sum.peak_sustained","sm__ops_path_tensor_src_fp64.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_int8.avg","sm__ops_path_tensor_src_int8.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.avg.peak_sustained","sm__ops_path_tensor_src_int8.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.avg.per_cycle_elapsed","sm__ops_path_tensor_src_int8.avg.per_second","sm__ops_path_tensor_src_int8.max","sm__ops_path_tensor_src_int8.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.max.peak_sustained","sm__ops_path_tensor_src_int8.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.max.per_cycle_elapsed","sm__ops_path_tensor_src_int8.max.per_second","sm__ops_path_tensor_src_int8.min","sm__ops_path_tensor_src_int8.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.min.peak_sustained","sm__ops_path_tensor_src_int8.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.min.per_cycle_elapsed","sm__ops_path_tensor_src_int8.min.per_second","sm__ops_path_tensor_src_int8.sum","sm__ops_path_tensor_src_int8.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.sum.peak_sustained","sm__ops_path_tensor_src_int8.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.sum.per_cycle_elapsed","sm__ops_path_tensor_src_int8.sum.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.avg","sm__ops_path_tensor_src_tf32_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.avg.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.max","sm__ops_path_tensor_src_tf32_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.max.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.min","sm__ops_path_tensor_src_tf32_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.min.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.sum","sm__ops_path_tensor_src_tf32_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.sum.per_second","sm__pipe_alu_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_tensor_subpipe_hmma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_subpipe_hmma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tensor_subpipe_imma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_subpipe_imma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__sass_inst_executed_op_ldgsts_cache_access.sum","sm__sass_inst_executed_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum.per_second","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts.sum","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","sm__sass_thread_inst_executed_op_dfma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.sum.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.sum.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.sum.peak_sustained","sm__throughput.avg.pct_of_peak_sustained_elapsed","sm__throughput.max.pct_of_peak_sustained_elapsed","sm__throughput.min.pct_of_peak_sustained_elapsed","sm__throughput.sum.pct_of_peak_sustained_elapsed","sm__warps_active.avg.pct_of_peak_sustained_active","sm__warps_active.avg.per_cycle_active","sm__warps_active.max.pct_of_peak_sustained_active","sm__warps_active.max.per_cycle_active","sm__warps_active.min.pct_of_peak_sustained_active","sm__warps_active.min.per_cycle_active","sm__warps_active.sum.pct_of_peak_sustained_active","sm__warps_active.sum.per_cycle_active","smsp__average_warp_latency_per_inst_issued.ratio","smsp__average_warps_active_per_inst_executed.ratio","smsp__average_warps_issue_stalled_barrier_per_issue_active.ratio","smsp__average_warps_issue_stalled_branch_resolving_per_issue_active.ratio","smsp__average_warps_issue_stalled_dispatch_stall_per_issue_active.ratio","smsp__average_warps_issue_stalled_drain_per_issue_active.ratio","smsp__average_warps_issue_stalled_lg_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_long_scoreboard_per_issue_active.ratio","smsp__average_warps_issue_stalled_math_pipe_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_membar_per_issue_active.ratio","smsp__average_warps_issue_stalled_mio_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_misc_per_issue_active.ratio","smsp__average_warps_issue_stalled_no_instruction_per_issue_active.ratio","smsp__average_warps_issue_stalled_not_selected_per_issue_active.ratio","smsp__average_warps_issue_stalled_selected_per_issue_active.ratio","smsp__average_warps_issue_stalled_short_scoreboard_per_issue_active.ratio","smsp__average_warps_issue_stalled_sleeping_per_issue_active.ratio","smsp__average_warps_issue_stalled_tex_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_wait_per_issue_active.ratio","smsp__branch_targets_threads_divergent","smsp__cycles_active.avg","smsp__cycles_active.max","smsp__cycles_active.min","smsp__cycles_active.sum","smsp__cycles_elapsed.avg","smsp__cycles_elapsed.avg.per_second","smsp__cycles_elapsed.max","smsp__cycles_elapsed.max.per_second","smsp__cycles_elapsed.min","smsp__cycles_elapsed.min.per_second","smsp__cycles_elapsed.sum","smsp__cycles_elapsed.sum.per_second","smsp__inst_executed.avg","smsp__inst_executed.max","smsp__inst_executed.min","smsp__inst_executed.sum","smsp__inst_executed_op_branch.avg","smsp__inst_executed_op_branch.max","smsp__inst_executed_op_branch.min","smsp__inst_executed_op_branch.sum","smsp__inst_executed_op_generic_atom_dot_alu.sum","smsp__inst_executed_op_generic_atom_dot_alu.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_generic_atom_dot_cas.sum","smsp__inst_executed_op_generic_atom_dot_cas.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_global_red.sum","smsp__inst_executed_op_global_red.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_ldgsts.sum","smsp__inst_executed_op_ldgsts.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_ldsm.sum","smsp__inst_executed_op_ldsm.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_shared_atom.sum","smsp__inst_executed_op_shared_atom.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_shared_stsm.sum","smsp__inst_executed_op_shared_stsm.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_atom_dot_alu.sum","smsp__inst_executed_op_surface_atom_dot_alu.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_atom_dot_cas.sum","smsp__inst_executed_op_surface_atom_dot_cas.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_ld.sum","smsp__inst_executed_op_surface_ld.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_red.sum","smsp__inst_executed_op_surface_red.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_st.sum","smsp__inst_executed_op_surface_st.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_texture.sum","smsp__inst_executed_op_texture.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_tma_ld.sum","smsp__inst_executed_op_tma_ld.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_tma_st.sum","smsp__inst_executed_op_tma_st.sum.pct_of_peak_sustained_elapsed","smsp__inst_issued.avg","smsp__inst_issued.max","smsp__inst_issued.min","smsp__inst_issued.sum","smsp__issue_active.avg.pct_of_peak_sustained_active","smsp__issue_active.avg.per_cycle_active","smsp__issue_active.max.pct_of_peak_sustained_active","smsp__issue_active.max.per_cycle_active","smsp__issue_active.min.pct_of_peak_sustained_active","smsp__issue_active.min.per_cycle_active","smsp__issue_active.sum.pct_of_peak_sustained_active","smsp__issue_active.sum.per_cycle_active","smsp__issue_inst0.avg.pct_of_peak_sustained_active","smsp__issue_inst0.max.pct_of_peak_sustained_active","smsp__issue_inst0.min.pct_of_peak_sustained_active","smsp__issue_inst0.sum.pct_of_peak_sustained_active","smsp__maximum_warps_avg_per_active_cycle","smsp__pcsamp_aggregated_passes","smsp__pcsamp_buffer_size_bytes","smsp__pcsamp_dropped_bytes","smsp__pcsamp_interval","smsp__pcsamp_interval_cycles","smsp__pcsamp_sample_count","smsp__pcsamp_warps_issue_stalled_barrier","smsp__pcsamp_warps_issue_stalled_barrier_not_issued","smsp__pcsamp_warps_issue_stalled_branch_resolving","smsp__pcsamp_warps_issue_stalled_branch_resolving_not_issued","smsp__pcsamp_warps_issue_stalled_dispatch_stall","smsp__pcsamp_warps_issue_stalled_dispatch_stall_not_issued","smsp__pcsamp_warps_issue_stalled_drain","smsp__pcsamp_warps_issue_stalled_drain_not_issued","smsp__pcsamp_warps_issue_stalled_lg_throttle","smsp__pcsamp_warps_issue_stalled_lg_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_long_scoreboard","smsp__pcsamp_warps_issue_stalled_long_scoreboard_not_issued","smsp__pcsamp_warps_issue_stalled_math_pipe_throttle","smsp__pcsamp_warps_issue_stalled_math_pipe_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_membar","smsp__pcsamp_warps_issue_stalled_membar_not_issued","smsp__pcsamp_warps_issue_stalled_mio_throttle","smsp__pcsamp_warps_issue_stalled_mio_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_misc","smsp__pcsamp_warps_issue_stalled_misc_not_issued","smsp__pcsamp_warps_issue_stalled_no_instructions","smsp__pcsamp_warps_issue_stalled_no_instructions_not_issued","smsp__pcsamp_warps_issue_stalled_not_selected","smsp__pcsamp_warps_issue_stalled_not_selected_not_issued","smsp__pcsamp_warps_issue_stalled_selected","smsp__pcsamp_warps_issue_stalled_selected_not_issued","smsp__pcsamp_warps_issue_stalled_short_scoreboard","smsp__pcsamp_warps_issue_stalled_short_scoreboard_not_issued","smsp__pcsamp_warps_issue_stalled_sleeping","smsp__pcsamp_warps_issue_stalled_sleeping_not_issued","smsp__pcsamp_warps_issue_stalled_tex_throttle","smsp__pcsamp_warps_issue_stalled_tex_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_wait","smsp__pcsamp_warps_issue_stalled_wait_not_issued","smsp__sass_average_branch_targets_threads_uniform.pct","smsp__sass_average_data_bytes_per_sector_mem_global_op_ld.max_rate","smsp__sass_average_data_bytes_per_sector_mem_global_op_ld.ratio","smsp__sass_average_data_bytes_per_sector_mem_global_op_st.max_rate","smsp__sass_average_data_bytes_per_sector_mem_global_op_st.ratio","smsp__sass_average_data_bytes_per_sector_mem_local_op_ld.max_rate","smsp__sass_average_data_bytes_per_sector_mem_local_op_ld.ratio","smsp__sass_average_data_bytes_per_sector_mem_local_op_st.max_rate","smsp__sass_average_data_bytes_per_sector_mem_local_op_st.ratio","smsp__sass_branch_targets_threads_divergent.avg","smsp__sass_branch_targets_threads_divergent.max","smsp__sass_branch_targets_threads_divergent.min","smsp__sass_branch_targets_threads_divergent.sum","smsp__sass_inst_executed_memdesc_explicit_evict_type","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_first","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_last","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_normal","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_normal_demote","smsp__sass_inst_executed_memdesc_explicit_missprop_evict_first","smsp__sass_inst_executed_memdesc_explicit_missprop_evict_normal","smsp__sass_inst_executed_op_dshared.sum","smsp__sass_inst_executed_op_dshared.sum.pct_of_peak_sustained_elapsed","smsp__sass_inst_executed_op_dshared_atom.sum","smsp__sass_inst_executed_op_dshared_ld.sum","smsp__sass_inst_executed_op_dshared_redas.sum","smsp__sass_inst_executed_op_dshared_st.sum","smsp__sass_inst_executed_op_dshared_stas.sum","smsp__sass_inst_executed_op_dshared_tma_red.sum","smsp__sass_inst_executed_op_dshared_tma_st.sum","smsp__sass_inst_executed_op_global_ld.sum","smsp__sass_inst_executed_op_global_st.sum","smsp__sass_inst_executed_op_local_ld.sum","smsp__sass_inst_executed_op_local_st.sum","smsp__sass_inst_executed_op_shared.sum","smsp__sass_inst_executed_op_shared.sum.pct_of_peak_sustained_elapsed","smsp__sass_inst_executed_op_shared_ld.sum","smsp__sass_inst_executed_op_shared_st.sum","smsp__sass_inst_executed_op_tma_ld.sum","smsp__sass_inst_executed_op_tma_red.sum","smsp__sass_inst_executed_op_tma_st.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts.sum.pct_of_peak_sustained_elapsed","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts_cache_access.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","smsp__sass_l1tex_m_xbar2l1tex_read_sectors_mem_global_op_ldgsts_cache_bypass.sum","smsp__sass_l1tex_m_xbar2l1tex_read_sectors_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.sum.per_cycle_elapsed","smsp__thread_inst_executed_per_inst_executed.ratio","smsp__thread_inst_executed_pred_on_per_inst_executed.ratio","smsp__warps_active.avg.peak_sustained","smsp__warps_active.avg.per_cycle_active","smsp__warps_active.max.peak_sustained","smsp__warps_active.max.per_cycle_active","smsp__warps_active.min.peak_sustained","smsp__warps_active.min.per_cycle_active","smsp__warps_active.sum.peak_sustained","smsp__warps_active.sum.per_cycle_active","smsp__warps_eligible.avg.per_cycle_active","smsp__warps_eligible.max.per_cycle_active","smsp__warps_eligible.min.per_cycle_active","smsp__warps_eligible.sum.per_cycle_active","thread_inst_executed","thread_inst_executed_true"
+"","","","","","","","","","","","","","thread","thread","thread","thread","","","","%","Kbyte","Gbyte","","","byte","","%","%/register","%/Kbyte","thread","thread","thread","%","thread","thread","thread","thread","thread","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","request","%","request","%","","%","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","ms","ms","ms","ms","","","","","block","block","block","block","","","","","%","%","%","%","%","","%","inst","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","","","","","","%","%","%","%","%","%","%","%","%","%","%","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","cycle","%","","Mhz","%","%","%","%","%","%","%","%","Mbyte","%","Gbyte/s","byte","%","byte/s","byte","%","byte/s","Mbyte","%","Gbyte/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector","%","sector/s","sector","%","sector/us","sector","%","sector","%","sector","%","sector","%","Gbyte","%","Gbyte/s","byte","%","byte/s","Gbyte","%","Gbyte/s","%","%","%","%","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector/ns","sector","%","sector","%","sector","%","sector","%","byte","%","byte/s","","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","%","%","cycle","%","%","%","%","%","%","%","%","%","%","%","%","%","","block","block","block","","","","","cluster","block","","","","","","","","","","","","%","%","block","block","block","block","block","","","","","","Mbyte","","","","","register/thread","register/thread","Kbyte","Kbyte/block","Kbyte/block","Kbyte/block","Kbyte/block","byte/block","SM","","","thread","","","","","","","","","%","sector","%","","","","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","sector","%","sector/ns","sector","sector","sector","sector","%","","Mhz","%","","Mhz","%","","Mhz","%","","Ghz","request","request","request","request","request","request","request","request","%","%","%","sector","%","sector/cycle","sector/cycle","sector","%","sector","%","sector","%","sector/ns","sector","%","sector/us","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/ns","%","sector/cycle","sector/cycle","sector","sector/us","sector","sector","%","sector/cycle","sector/cycle","%","%","sector","%","sector/ns","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/s","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/ns","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/s","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/us","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","%","%","%","%","%","%","%","%","%","%","%","sector","%","sector/s","sector","%","sector/s","","","sectors","sectors","sectors","sectors","sectors","sectors","","","","","","","","","","","","","","","","","","","","","","","","block","","cycle","%","%","%","cycle","%","inst/cycle","%","%","%","warp","warp","","Mbyte","","","us","sample","","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","pass","pass","","","","","inst","inst","inst","inst","","","","","inst","inst","inst","inst","inst","inst","inst","","","","","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","cycle","%","cycle","%","cycle","%","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","inst/cycle","%","inst/cycle","%","inst/cycle","%","inst/cycle","%","%","%","%","%","%","%","%","%","%","%","%","warp","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","inst","inst","byte","%","byte/s","","%","","","sector","%","sector","%","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","%","%","%","%","%","warp","%","warp","%","warp","%","warp","cycle","cycle","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","branches","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","inst","inst","inst","inst","inst","inst","inst","inst","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","inst","inst","inst","%","","%","","%","","%","","%","%","%","%","warp","pass","Mbyte","byte","","cycle","","warp","warp","branches","branches","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","inst","inst","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","%","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","branches","branches","branches","branches","","inst","inst","inst","inst","inst","inst","inst","%","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","%","inst","inst","inst","inst","inst","","%","","%","sector","%","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","","","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","inst","inst"
+"0","73","python3.12","127.0.0.1","void at::vectorized_elementwise_kernel<4, at::BinaryFunctor<float, float, float, at::MulFunctor<float>>, std::array<char *, 3>>(int, T2, T3)","1","7","(128, 1, 1)","(1, 1, 1)","0","12.1","0","1","3802","2078","3874.000000","1694","6144","34334763.95","0","0.000000","1.024000","3.326180","0","0","0","692","5270.000000","10876.000000","1.959000","192.000000","12288.000000","12288.000000","0.075665","0","0.000000","0","0.128805","0","37","489","0","0","nan","0","432","1","0","1","1","1","0","1","0","1","0","443","2418000","1","12","1","0","1","1","1","No-CC","1","1","1","0","0","NVIDIA GB10","0","0","256","4","1","1","256","1","0","0","100","1","772935902","11794","32000","16","0","4","2500","16","161","4318","0","1","0","0","1","0","1","443","1","1","1","25165824","24","24","24","1","1","134217728","1024","1024","64","24","2418000","2147483647","65535","65535","4","1","8533000","18874368","2147483647","65536","65536","255","49152","101376","102400","1024","1536","48","12","2048","32768","32768","65536","32768","2048","32768","131072","16384","16384","16384","2046","32768","32768","2048","32768","268435456","32768","131072","32768","32768","65536","32768","2048","32768","65000","2097120","131072","32768","32768","131072","16384","32768","16384","8192","16384","8192","2046","32768","32768","4","8533000","1","1","0","0","0","0","48","6","4","1","0","1","1","1","0","15","1","0","1024","12","64","1","1","512","0","1","512","32","65536","130661769216","1","1","1","32","4.000000","0.006289","30.000000","0.047170","672","0.528302","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","63600.000000","8.530043","0.264263","0.861043","0.014001","0.264263","0.146132","0.448022","0.014001","0.146132","0.251575","0.918239","0.056604","0.251575","0","0","0","0","0.264263","0.918239","0.056604","0.264263","0.007456","0.007456","0.007456","0.007456","0","0","0","0","0.000000","0.000000","0.000000","0.000000","0","0","0","0","0.026468","1.270440","0.000000","0.026468","53.623188","69","0.018082","489","46.625000","2238.000000","0.000000","2238.000000","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","763200.000000","102.360515","0","0","0","0","0","0.000008","0.000393","0.000000","0.000008","0.000037","0.001769","0.000000","0.000037","0.009565","0.459119","0.000000","0.009565","70","0.009172","0","0","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0.000786","0.037736","0.000000","0.000786","2.000000","0.000262","48","0.268240","0.019130","0.918239","0.000000","0.019130","0.012972","0.031447","0.012579","0.012972","0.000032","0.000131","0.004292","0","0","0","0","0","0","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","1.000000","0.000131","0","0","0","0","0","0","0.000000","0.000262","0.008584","0","0","0","0.000000","0.000000","0.000000","0.000262","0.012579","0.000000","0.000262","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","2.000000","0.000262","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2","0.000262","0","0","1","0.000131","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2","0.000262","0","0","1","0.000131","0","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0","0","0.000000","0","0.000000","0","0","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","0.000000","2.000000","0","0","0","1.000000","0","1.000000","0.000000","0.000000","0","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.056604","0.056604","0.056604","0.056604","19.302949","0.056604","313.136729","0.918239","19.302949","0.056604","19.302949","0.056604","0","128","1","1","128","0","0","0","0","8","PolicySpread","0","1","0","CachePreferNone","1","1","1","1","1","","0","0","24.000000","24.000000","12.000000","32.000000","12.000000","336","2540","0","5264","956","4.718592","0","0","0","0","40","40.000000","32.768000","1.024000","1.024000","1.024000","0.000000","0","48","1024","7","128","24","all","0","0","0","0","0","0.00","100","0","0","0","0","0","1253.875000","3557.000000","149.000000","20062.000000","14285.000000","1.915907","14285.000000","1.915907","14285.000000","1.915907","228560.000000","30.654506","0","0","0","0","0","0","0","0","0.169758","0.434022","0.000000","0.169758","185.000000","0.080942","0.024812","0","0","0","0","0.169540","2","6.496446","0.700035","2","26.824034","0.007000","2","0.268240","0.169540","32","0.103943","301.000000","176.000000","3.000000","0","0","2.000000","0","1.000000","48.758278","73.684211","1.995012","75.500000","0.264263","2.000000","0.005285","246.000000","0.861043","2.000000","0.007000","1208.000000","0.264263","0.162017","401.000000","0.175446","53.782189","0","8.000000","0","0","0","57.000000","284.000000","0","0","0","0","549.000000","193.000000","589.000000","535.000000","0.000000","0.000000","0.000000","0.308015","1.000000","0.003080","704.000000","94.420601","533.000000","171.000000","0.000656","2.000000","0.000013","0.003500","0.000000","3.000000","0.000656","0.000402","0","0","0","0","0.187500","1.000000","0.000000","3.000000","0","0","0","0","0","0","0.000000","3.000000","0.000000","0.187500","1.000000","0.000000","3.000000","0","0","0","0","1.000000","0","0","0","0","0","0","1.000000","0","0","0","0","0","0","0","0","0","0","0","0.000438","2.000000","0.000009","2.000000","0.000438","0.000268","0","0","0","0","0","0","0.000000","2.000000","0.000000","2.000000","0","1.000000","0","0","0","0","0","0","0.000438","1.000000","0.000004","1.000000","0.000438","0.134120","0","0","0","0","0","0","0","1.000000","0","1.000000","0.139132","0.469023","0.000000","0.139132","0.264263","0.861043","0.014001","0.264263","0.251575","0.602030","0.014001","0.251575","0","0","0","0","0","0","0","0","3","0","0","3","3","0","0","1","1","1","1","0","0","0","0","0","0","0","0","1","0","0","0","0","0","0","0","0","0","0.000000","1.52","0.125000","","","","641.229167","0.000055","0.154119","0.000000","0.000117","0.000023","0.286469","13.750532","0","4.194304","1","1","3.000000","0","2","25.185856","25.185856","25.185856","957.062541","25.185856","25.185856","25.185856","957.062541","38.000000","0","1","1","1","1","2.000000","1.000000","0.000000","0.000000","489","489","489","489","0","no data","no data","0","0","0.000000","0.000000","8331","8331","8331","8331","46.625000","2238.000000","0.000000","2238.000000","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","763200.000000","102.360515","0.000000","8.000000","0.001048","0.000000","73.333333","120.000000","0.031447","0.016018","0.218499","0.000641","0.768868","10.487936","0.030755","0.000000","0.000000","0.000000","0.016018","10.487936","0.030755","9.740840","0.028564","467.560322","1.371069","0.000000","0.000000","9.740840","0.028564","1.843164","0.005405","88.471850","0.259434","0.000000","0.000000","1.843164","0.005405","0.290438","0.000852","13.941019","0.040881","0.000000","0.000000","0.290438","0.000852","0.003407","0.163522","0.000000","0.003407","1.351653","0.003964","64.879357","0.190252","0.000000","0.000000","1.351653","0.003964","0.357462","0.001048","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","6.523682","0.019130","313.136729","0.918239","0.000000","0.000000","6.523682","0.019130","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0.016051","0.218945","0.770440","10.509383","0.000000","0.000000","0.016051","10.509383","0.016051","0.770440","0.000000","0.016051","0","0","0","0","0.016051","0.770440","0.000000","0.016051","48.000000","100.000000","0.028564","1.371069","0.000000","0.028564","0.001048","0.050314","0.000000","0.001048","0.002588","0.124214","0.000000","0.002588","0.015898","0.763103","0.000000","0.015898","0.000393","0.018868","0.000000","0.000393","0.000393","0.018868","0.000000","0.000393","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","49152","104817167381974.25","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","49152","104817167381974.25","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","49152","104817167381974.25","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","512","1091845493562.23","0","0","0","0","512","1091845493562.23","0","0","0","0","512","1091845493562.23","0","0","0","0","24576","52408583690987.12","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","49152","104817167381974.25","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0.000000","4096","8734763948497.85","0","0","0","0.000000","4096","8734763948497.85","0","0","0","0.000000","4096","8734763948497.85","0","0","0","0.000000","196608","419268669527897","0","0","0","0","8192","17469527896995.71","0","0","0","0","8192","17469527896995.71","0","0","0","0","8192","17469527896995.71","0","0","0","0","393216","838537339055794","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","3.51","7478393791.52","0","0","0","0","3.51","7478393791.52","0","0","0","0","3.51","7478393791.52","0","0","0","0","168.33","358962901993.06","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","2048","4367381974248.93","0","0","0","0","98304","209634334763948.50","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","4096","8734763948497.85","0","0","0","0","196608","419268669527897","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","1024","2183690987124.46","0","0","0","0","49152","104817167381974.25","0","0","1.843164","0.005405","88.471850","0.259434","0.000000","0.000000","1.843164","0.005405","0.006420","0.308176","0.000000","0.006420","0.804290","0.002358","38.605898","0.113208","0.000000","0.000000","0.804290","0.002358","0.015789","0.757862","0.000000","0.015789","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0","0","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","2.000000","2.000000","96.000000","128.000000","128.000000","128.000000","6144.000000","64.000000","64.000000","64.000000","3072.000000","0.028564","1.371069","0.000000","0.028564","32.530906","15.614835","1561.483467","749.512064","0.000000","0.000000","32.530906","749.512064","12.938776","12.965235","0.000000","0.381633","0.008163","0.016327","0","1.822449","0.022449","0","0.000000","0.012245","90.138776","0.000000","1.000000","2.073469","0.000000","0","2.055102","1","33.052083","2259.000000","0.000000","6346.000000","15900.000000","2.132511","15903.000000","2.132913","15897.000000","2.132108","3052800.000000","409.442060","2.546875","132.000000","0.000000","489.000000","0.192708","13.000000","0.000000","37.000000","0","0","0","0","0","0","0","0","0.000000","0.000000","0","0","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","0.000000","2.552083","133.000000","0.000000","490.000000","7.721399","0.08","402.395210","4.02","0.000000","0","7.721399","14.83","92.278601","6432.272298","0.000000","92.278601","12.000000","2","33.554432","0","5","1024.000000","16","0","0","0","0","0","0","0","0","0","0","8","8","0","0","0","0","0","0","0","0","6","6","0","0","0","0","2","2","0","0","0","0","0","0","94.736842","32.000000","4.000000","32.000000","4.000000","32.000000","0.000000","32.000000","0.000000","0.005208","1.000000","0","1.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","1.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0.000000","0","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.000000","0.000000","0.000000","0.000000","0.000294","0.014088","0","0.056352","0","0","0","0","0.000168","0.008050","0","0.032201","0","0","0","0","31.15","17.04","12.000000","0.999055","12.000000","69.284589","12.000000","0.000000","2304.000000","191.818468","0.077214","4.023952","0.000000","14.825087","15233","8331"
+"1","73","python3.12","127.0.0.1","cutlass3x_sm120_bstensorop_s16864gemm_block_scaled_ue4m3xe2m1_ue4m3xe2m1_f32_bf16_bf16_128x128x256_1x1x1_0_tnn_align32_o_vs16_bias_bf16_relu_stream_k","1","7","(384, 1, 1)","(42, 296, 1)","0","12.1","0","1","56167","52394","354166.000000","7598","6144","1620772988.98","696768","100.000000","1.024000","466.239645","294","44556288","0","625","316.000000","4225.000000","4.975000","192.000000","12288.000000","12288.000000","0.033180","0","3.443262","0","0.000000","0","42996614","1295861238","576","696768","1","1","432","1","0","1","1","1","0","1","0","1","0","443","2418000","1","12","1","0","1","1","1","No-CC","1","1","1","0","0","NVIDIA GB10","0","0","256","4","1","1","256","1","0","0","100","1","772935902","11794","32000","16","0","4","2500","16","161","4318","0","1","0","0","1","0","1","443","1","1","1","25165824","24","24","24","1","1","134217728","1024","1024","64","24","2418000","2147483647","65535","65535","4","1","8533000","18874368","2147483647","65536","65536","255","49152","101376","102400","1024","1536","48","12","2048","32768","32768","65536","32768","2048","32768","131072","16384","16384","16384","2046","32768","32768","2048","32768","268435456","32768","131072","32768","32768","65536","32768","2048","32768","65000","2097120","131072","32768","32768","131072","16384","32768","16384","8192","16384","8192","2046","32768","32768","4","8533000","1","1","0","0","0","0","48","6","4","1","0","1","1","1","0","15","1","0","1024","12","64","1","1","512","0","1","512","32","65536","130661769216","1","1","1","32","2798544.000000","0.591359","9028.000000","0.001908","10712","0.001132","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","473239351.000000","8.595390","23.974686","23.993901","23.961532","23.974686","12.559269","12.628962","12.489907","12.559269","23.598833","23.616927","23.583289","23.598833","0","0","0","0","23.974686","23.993901","23.961532","23.974686","55.057344","55.057344","55.057344","55.057344","12384","12384","12384","12384","12384.000000","12384.000000","12384.000000","12384.000000","12432","12432","12432","12432","0.070323","0.070326","0.070310","0.070323","99.721630","1991232","0.070128","1259175041","118776167.583333","118805817.000000","118739425.000000","5701256044.000000","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","5678872212.000000","103.144682","44813290","0","44566514","0","250754","5.324639","5.324688","5.324632","5.324639","1.793456","1.793457","1.793455","1.793456","12.737439","12.738040","12.736932","12.737439","719808603","12.675203","0","0","713157733","12.558087","3440663","0.060587","0","0","0","0","0","0","0","0","11.787449","11.787449","11.787449","11.787449","697152.000000","0.012276","48","12.662289","7.999717","7.999717","7.999717","7.999717","3.560743","3.563408","3.559783","3.560743","406.757376","0.223833","7.387886","0","0","0","0","0","0","406.683648","0.223792","7.386547","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","12708864.000000","0.223792","230.829587","2304.000000","0.000041","0","0","0","0","0","0","25.664425","14.122756","466.139902","0","0","0","25.664422","14.122755","466.139847","14.122756","14.122756","14.122756","14.122756","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","802013184.000000","14.122755","14.566870","96.000000","0.000002","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000017","0","0","0","0.000000","696192","0.012259","576","0.000010","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000017","0","0","0","0.000000","696192","0.012259","576","0.000010","0","0","0","0","0","0","0","0","0","0","99.979413","0","90.000000","0","0","100.000000","97.916667","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960.000000","864.000000","96.000000","0","0","0","0.000000","0","0.000000","696192.000000","696192.000000","0","2304.000000","2256.000000","48.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.011638","0.011678","0.011589","0.011638","14.067309","14.122756","14.067309","14.122756","14.067309","14.122756","14.067309","14.122756","8","384","1","1","384","0","0","0","0","8","PolicySpread","0","1","0","CachePreferNone","1","42","296","1","12432","","0","0","3.000000","24.000000","1.000000","1.000000","4.000000","300","157","0","2028","2388","4.718592","0","0","0","0","168","168.000000","102.400000","89.088000","89.088000","1.024000","88.064000","0","48","1024","7","4773888","24","all","0","0","0","0","0","259","100","0","0","0","1.00","0","107113733.250000","107135103.000000","107088888.000000","1713819732.000000","106226770.000000","1.929384","106226770.000000","1.929384","106226770.000000","1.929384","1699628320.000000","30.870147","0","0","0","0","0","0","0","0","17.584039","17.594977","17.573523","17.584039","367742418.000000","21.636637","6.679262","0","0","0","0","23.598833","2","910.624306","23.616927","2","911.322529","23.583289","2","910.024519","23.598833","32","14.569989","203788027.000000","2832.000000","203681184.000000","0","0","200503392.000000","0","3177792.000000","53.315402","54.157122","0.390314","50935069.437500","23.974686","2.000000","0.479494","50975893.000000","23.993901","50907123.000000","23.961532","814961111.000000","23.974686","14.802042","12769216.000000","0.751295","231.925754","0","8.000000","0","0","0","63635.000000","209527.000000","0","0","0","0","434448305.000000","380287608.000000","434499795.000000","380523451.000000","802013280.000000","5.898446","14.566872","0.000666","1.000000","0.000007","11328.000000","0.205749","8632.000000","2688.000000","23.967724","2.000000","0.479354","23.982646","23.954898","814724448.000000","23.967724","14.797743","0","0","0","0","23767776.125000","23784480.000000","23750968.000000","380284418.000000","0","0","0","0","0","0","434439826.000000","380285022.000000","434440114.000000","23767813.875000","23784508.000000","23750980.000000","380285022.000000","0","0","0","0","1.000000","0","0","0","0","0","0","1.000000","0","0","0","0","0","0","0","0","0","0","0","23.593784","2.000000","0.471876","802013280.000000","23.593784","14.566872","0","0","0","0","0","0","434439462.000000","367573666.000000","434439462.000000","367573666.000000","0","1.000000","0","0","0","0","0","0","0.747879","1.000000","0.007479","12711168.000000","0.747879","230.871435","0","0","0","0","0","0","0","12711168.000000","0","12711168.000000","11.990160","12.008335","11.982277","11.990160","23.974686","23.993901","23.961532","23.974686","11.920534","11.945496","11.914528","11.920534","0","0","0","0","0","0","0","0","960","716332272","671775984","960","960","698496","0","1","1","1","1","0","0","0","0","0","0","0","0","1","0","0","0","0","0","0","0","0","0","32.000000","15070513.85","13945712.000000","","","","116848615.062500","1.442512","1048.198735","24.916459","0.716718","0.027825","45312.875305","2175018.014626","0","28.770304","1","1","12.000000","0","2","25.185856","25.185856","25.185856","957.062541","25.185856","25.185856","25.185856","957.062541","38.000000","0","1","1","1","1","960.000000","0.000000","696192.000000","576.000000","1259175041","1259175041","1259175041","1259175041","696768","no data","no data","696192","576","89236896.000000","432.000000","31879588511","31879588511","31879588511","31879588511","118776167.583333","118805817.000000","118739425.000000","5701256044.000000","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","5678872212.000000","103.144682","99.622652","1476887.000000","0.026007","1471314.000000","99.945294","51933628.000000","1.829012","5.704747","0.227294","0.228190","5.705824","0.227337","0.228233","5.703584","0.227248","0.228143","5.704747","10.910112","10.953115","1.367979","1.373371","1.370839","1.376242","1.364895","1.370275","1.367979","1.373371","1.447909","1.453616","1.448026","1.453733","1.447743","1.453449","1.447909","1.453616","0.057268","0.057494","0.057434","0.057660","0.057106","0.057331","0.057268","0.057494","0.229003","0.229722","0.228182","0.229003","0.159191","0.159818","0.159191","0.159818","0.159191","0.159818","0.159191","0.159818","0.000000","0.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","7.968309","7.999717","7.968309","7.999717","7.968309","7.999717","7.968309","7.999717","25.008546","25.107119","0","0","0","0","0","0","0","0","0","0","0.157014","0.157633","0.157014","0.157633","0.157014","0.157633","0.157014","0.157633","1.046402","1.050527","1.046402","1.050527","1.046402","1.050527","1.046402","1.050527","0.000218","0.000219","0.000218","0.000219","0.000218","0.000219","0.000218","0.000219","0","0","0","0","0","0","0","0","5.748362","0.229032","5.749550","0.229079","5.747026","0.228978","5.748362","10.993524","25.107119","25.107119","25.107119","25.107119","0","0","0","0","5.748362","5.749550","5.747026","5.748362","12.000000","25.000000","7.999717","7.999717","7.999717","7.999717","0.026006","0.026006","0.026006","0.026006","2.955628","2.955628","2.955628","2.955628","3.159392","3.160349","3.158360","3.159392","1.329433","1.334115","1.324826","1.329433","1.329441","1.334122","1.324833","1.329441","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","49152","105620153872442.52","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","49152","105620153872442.52","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","49152","105620153872442.52","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","512","1100209936171.28","0","0","0","0","512","1100209936171.28","0","0","0","0","512","1100209936171.28","0","0","0","0","24576","52810076936221.26","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","49152","105620153872442.52","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","121668370432","25.107119","4096","8801679489370.21","1028.39","2209848161800.18","121668370432","25.107119","4096","8801679489370.21","1028.39","2209848161800.18","121668370432","25.107119","4096","8801679489370.21","1028.39","2209848161800.18","5840081780736","25.107119","196608","422480615489770.06","49362.60","106072711766408.50","0","0","8192","17603358978740.42","0","0","0","0","8192","17603358978740.42","0","0","0","0","8192","17603358978740.42","0","0","0","0","393216","844961230979540.12","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","3.51","7535684494.32","0","0","0","0","3.51","7535684494.32","0","0","0","0","3.51","7535684494.32","0","0","0","0","168.33","361712855727.54","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","2048","4400839744685.10","0","0","0","0","98304","211240307744885.03","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","4096","8801679489370.21","0","0","0","0","196608","422480615489770.06","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","1024","2200419872342.55","0","0","0","0","49152","105620153872442.52","0","0","1.447899","1.453606","1.448020","1.453728","1.447714","1.453420","1.447899","1.453606","2.476614","2.476884","2.476247","2.476614","0.239030","0.239972","0.239497","0.240441","0.238574","0.239514","0.239030","0.239972","0.723077","0.723126","0.723030","0.723077","0","0","0","0","0","0","0","0","25.008546","25.107119","25.008546","25.107119","25.008546","25.107119","25.008546","25.107119","25.008546","25.107119","0","0","0.157014","0.157633","0.157014","0.157633","0.157014","0.157633","0.157014","0.157633","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","2.000000","2.000000","96.000000","128.000000","128.000000","128.000000","6144.000000","64.000000","64.000000","64.000000","3072.000000","25.107119","25.107119","25.107119","25.107119","20.749477","9.959749","20.755541","9.962660","20.742386","9.956345","20.749477","478.067958","43.772834","44.107494","0.206873","0.162419","0.067445","0.000002","0","0.653241","3.781903","0","0.022919","0.018141","0.089313","0.278872","1.000002","0.105348","33.715651","0","3.744846","12480","118853673.520833","118890151.000000","118818410.000000","22819905316.000000","118309837.750000","2.148848","118334635.000000","2.149298","118285340.000000","2.148403","22715488848.000000","412.578726","6749277.281250","7300353.000000","6518321.000000","1295861238.000000","223940.697917","308958.000000","174699.000000","42996614.000000","0","0","0","0","0","0","0","0","133668864.000000","0.588448","0","0","1591296.000000","0.007005","0","0","0","0","0","0","0","0","0","0","0","0","2784768.000000","0.012259","99456.000000","0.000438","6800878.151042","7388883.000000","6547928.000000","1305768605.000000","5.722060","0.06","6.216790","0.06","5.509235","0.06","5.722060","10.99","94.277940","93.813901","94.461095","94.277940","3.000000","2","33.554432","0","5","1024.000000","3262109","32325","30631","11408","9613","5261","2048","1","1","0","0","59347","55050","294725","248195","0","0","2176","2043","1755","1465","7581","6623","21812","0","77232","0","7732","6612","2450234","2418252","0","0","290520","282914","99.937098","32.000000","4.000000","32.000000","0.000000","32.000000","4.000000","32.000000","1.000000","65.000000","307.000000","0","12480.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960.000000","0.000000","696192.000000","576.000000","224497488.000000","0.988301","89236896.000000","432.000000","2784768.000000","0","99456.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.008967","0.010594","0.007720","1.721631","0.000000","0.000000","0","0.000000","0","0","0","0","0.000000","0.000000","0","0.000000","0","0","0","0","31.18","25.44","12.000000","2.504708","12.000000","3.006371","12.000000","2.003303","2304.000000","480.903877","0.073179","0.078745","0.070831","14.050374","39306980848","31879588511"
diff --git a/benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep b/benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep
new file mode 100644
index 0000000..a28664a
Binary files /dev/null and b/benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep differ
diff --git a/benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json b/benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json
new file mode 100644
index 0000000..8ed4f86
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json
@@ -0,0 +1,1733 @@
+{
+  "mode": "profile",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 12,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 32,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "split_k": 1,
+    "reduction_scheme": 0,
+    "custom_option": 0,
+    "cta_swizzle": 0
+  },
+  "profile": {
+    "target": "baseline",
+    "selected_config": null,
+    "supplied_workspace_bytes": 33554432,
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json b/benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json
new file mode 100644
index 0000000..36ad8cf
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json
@@ -0,0 +1,2014 @@
+{
+  "mode": "block-gate",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 12,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 32,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "split_k": 1,
+    "reduction_scheme": 0,
+    "custom_option": 0,
+    "cta_swizzle": 0
+  },
+  "block_gate": [
+    {
+      "block": 0,
+      "only_monkeypatched_method": "block.mlp.fc2.forward_swiglu",
+      "accepted_gate_and_residual_path_preserved": true,
+      "candidate_vs_baseline": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147",
+        "expected_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147"
+      },
+      "baseline_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147",
+        "expected_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147"
+      },
+      "candidate_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147",
+        "expected_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147"
+      },
+      "timing": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            457.58404541015625,
+            455.8675537109375,
+            455.042724609375,
+            456.5163879394531,
+            456.7074279785156,
+            459.7113037109375,
+            456.3467102050781,
+            459.0717468261719,
+            457.6736755371094,
+            458.6617736816406,
+            458.61083984375,
+            456.4378662109375,
+            457.4807434082031,
+            458.178466796875,
+            457.4196472167969,
+            458.90869140625,
+            458.4776306152344,
+            458.2261962890625,
+            458.10919189453125,
+            460.418701171875
+          ],
+          "p50_ms": 457.8914337158203,
+          "p95_ms": 459.7466735839844,
+          "mean_ms": 457.7725662231445
+        },
+        "candidate": {
+          "samples_ms": [
+            428.5652160644531,
+            425.1132507324219,
+            420.88983154296875,
+            420.71435546875,
+            419.1617736816406,
+            421.9552307128906,
+            419.37896728515625,
+            419.3441467285156,
+            419.79351806640625,
+            419.0124206542969,
+            420.5487976074219,
+            420.9779052734375,
+            420.5229187011719,
+            420.88104248046875,
+            420.3728942871094,
+            420.11834716796875,
+            420.3209533691406,
+            423.73919677734375,
+            419.7261657714844,
+            423.4720458984375
+          ],
+          "p50_ms": 420.5358581542969,
+          "p95_ms": 425.2858489990234,
+          "mean_ms": 421.2304489135742
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147",
+          "expected_sha256": "6bdb03e2bdf7733fde92b2cf5c78a373e9818cb5ec9d213d873a632fad0af147"
+        }
+      },
+      "supplied_workspace_bytes": 0
+    },
+    {
+      "block": 24,
+      "only_monkeypatched_method": "block.mlp.fc2.forward_swiglu",
+      "accepted_gate_and_residual_path_preserved": true,
+      "candidate_vs_baseline": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75",
+        "expected_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75"
+      },
+      "baseline_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75",
+        "expected_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75"
+      },
+      "candidate_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75",
+        "expected_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75"
+      },
+      "timing": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            458.4346008300781,
+            459.98040771484375,
+            457.2791748046875,
+            456.69989013671875,
+            458.1690368652344,
+            459.9689025878906,
+            459.0711975097656,
+            461.275390625,
+            458.7782897949219,
+            460.40740966796875,
+            459.8872985839844,
+            461.1615905761719,
+            459.39276123046875,
+            459.6435546875,
+            460.915283203125,
+            461.53076171875,
+            460.4534912109375,
+            460.2518615722656,
+            458.4383544921875,
+            461.34747314453125
+          ],
+          "p50_ms": 459.9281005859375,
+          "p95_ms": 461.3566375732422,
+          "mean_ms": 459.6543365478516
+        },
+        "candidate": {
+          "samples_ms": [
+            418.2376708984375,
+            421.28668212890625,
+            417.4606018066406,
+            420.7315368652344,
+            421.40478515625,
+            419.4402770996094,
+            417.5429382324219,
+            419.78125,
+            419.68603515625,
+            421.91644287109375,
+            422.23394775390625,
+            420.081298828125,
+            419.1646728515625,
+            421.63360595703125,
+            422.85125732421875,
+            418.6084899902344,
+            420.3404235839844,
+            421.3265075683594,
+            422.9977111816406,
+            420.1449279785156
+          ],
+          "p50_ms": 420.24267578125,
+          "p95_ms": 422.8585800170899,
+          "mean_ms": 420.3435531616211
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75",
+          "expected_sha256": "d8f5c6335e823dc4e3ed8cdd34d90dda0d69297ea47de9b328962dfae00c9f75"
+        }
+      },
+      "supplied_workspace_bytes": 0
+    },
+    {
+      "block": 49,
+      "only_monkeypatched_method": "block.mlp.fc2.forward_swiglu",
+      "accepted_gate_and_residual_path_preserved": true,
+      "candidate_vs_baseline": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6",
+        "expected_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6"
+      },
+      "baseline_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6",
+        "expected_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6"
+      },
+      "candidate_vs_traversal": {
+        "bf16_exact": true,
+        "different_elements": 0,
+        "max_abs": 0.0,
+        "mean_abs": 0.0,
+        "actual_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6",
+        "expected_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6"
+      },
+      "timing": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            455.2886962890625,
+            456.0704650878906,
+            455.62127685546875,
+            459.4961242675781,
+            456.5694580078125,
+            457.70477294921875,
+            458.9316101074219,
+            458.7506103515625,
+            457.3757629394531,
+            456.0811462402344,
+            458.1807861328125,
+            457.0354919433594,
+            459.8126220703125,
+            457.84747314453125,
+            458.5899963378906,
+            459.2362060546875,
+            460.5073547363281,
+            459.9343566894531,
+            457.9271240234375,
+            458.1864318847656
+          ],
+          "p50_ms": 458.053955078125,
+          "p95_ms": 459.9630065917969,
+          "mean_ms": 457.95738830566404
+        },
+        "candidate": {
+          "samples_ms": [
+            414.79840087890625,
+            421.3484191894531,
+            416.3827209472656,
+            416.5849914550781,
+            419.1302490234375,
+            414.7456970214844,
+            419.312255859375,
+            415.2906494140625,
+            416.9346008300781,
+            416.1675720214844,
+            418.51458740234375,
+            417.76995849609375,
+            420.13006591796875,
+            418.7623596191406,
+            416.8761901855469,
+            418.8280029296875,
+            417.02777099609375,
+            417.6993408203125,
+            416.9612121582031,
+            419.1832580566406
+          ],
+          "p50_ms": 417.3635559082031,
+          "p95_ms": 420.190983581543,
+          "mean_ms": 417.62241516113284
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6",
+          "expected_sha256": "3723ebf82eb1e279cc004832b68d071422a45dcc74adac93010e5fde33b715d6"
+        }
+      },
+      "supplied_workspace_bytes": 0
+    }
+  ],
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json b/benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json
new file mode 100644
index 0000000..163a75b
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json
@@ -0,0 +1,2133 @@
+{
+  "mode": "sweep",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 12,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 64,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 64,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 64,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 64,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 64,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 64,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "custom_option": 0,
+    "cta_swizzle": 0,
+    "inner_shape": null,
+    "cluster_shape": null,
+    "split_k": 1,
+    "reduction_scheme": 0
+  },
+  "candidates": [
+    {
+      "selected_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "split_k": -2,
+        "reduction_scheme": 2,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "required_workspace_bytes": 0,
+        "waves": 1.0,
+        "state": 0,
+        "api_status": 0,
+        "valid": true,
+        "capabilities": {
+          "split_k_support": 1,
+          "reduction_scheme_mask": 6,
+          "cta_swizzle_support": 0,
+          "custom_option_max": 0,
+          "strided_batch_support": 1,
+          "out_of_place_result_support": 1,
+          "tile_ids": [
+            20
+          ],
+          "stages_ids": [
+            37
+          ],
+          "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+        }
+      },
+      "supplied_workspace_bytes": 67108864,
+      "required_workspace_bytes": 0,
+      "checked": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "split_k": -2,
+        "reduction_scheme": 2,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "required_workspace_bytes": 0,
+        "waves": 1.0,
+        "state": 0,
+        "api_status": 0,
+        "valid": true,
+        "capabilities": {
+          "split_k_support": 1,
+          "reduction_scheme_mask": 6,
+          "cta_swizzle_support": 0,
+          "custom_option_max": 0,
+          "strided_batch_support": 1,
+          "out_of_place_result_support": 1,
+          "tile_ids": [
+            20
+          ],
+          "stages_ids": [
+            37
+          ],
+          "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+        }
+      },
+      "fc2_only": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            52.528385162353516,
+            53.54975891113281,
+            52.53424072265625,
+            52.453086853027344,
+            52.415489196777344,
+            53.50998306274414,
+            52.42780685424805,
+            52.526817321777344,
+            52.41756820678711,
+            52.83504104614258,
+            54.08259201049805,
+            52.47983932495117,
+            52.43289566040039,
+            52.59030532836914,
+            52.4370231628418,
+            52.52783966064453,
+            52.4318733215332,
+            52.5145263671875,
+            52.424705505371094,
+            52.57904052734375
+          ],
+          "p50_ms": 52.52067184448242,
+          "p95_ms": 53.57640056610108,
+          "mean_ms": 52.68494091033936,
+          "dense_tflop_s_p50": 110.9669507956279
+        },
+        "candidate": {
+          "samples_ms": [
+            52.557823181152344,
+            53.029598236083984,
+            52.549888610839844,
+            52.55766296386719,
+            52.51689529418945,
+            53.11779022216797,
+            54.02726364135742,
+            52.74710464477539,
+            52.530174255371094,
+            52.540096282958984,
+            52.56089782714844,
+            52.54924774169922,
+            52.527103424072266,
+            52.53104019165039,
+            53.064640045166016,
+            53.72809600830078,
+            53.41603088378906,
+            52.76732635498047,
+            52.54451370239258,
+            53.056480407714844
+          ],
+          "p50_ms": 52.55936050415039,
+          "p95_ms": 53.74305438995361,
+          "mean_ms": 52.84598369598389,
+          "dense_tflop_s_p50": 110.88526862612385
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f",
+          "expected_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+        }
+      },
+      "accepted_producer_plus_fc2": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            75.45340728759766,
+            75.004638671875,
+            76.25904083251953,
+            75.5995864868164,
+            75.02406311035156,
+            75.02108764648438,
+            74.97523498535156,
+            76.13423919677734,
+            76.12108612060547,
+            76.02950286865234,
+            76.11692810058594,
+            75.5208969116211,
+            75.1124496459961,
+            75.58025360107422,
+            74.99574279785156,
+            76.13922882080078,
+            76.13350677490234,
+            76.16307067871094,
+            76.70272064208984,
+            75.0445785522461
+          ],
+          "p50_ms": 75.58992004394531,
+          "p95_ms": 76.28122482299804,
+          "mean_ms": 75.65656318664551,
+          "dense_tflop_s_p50": 77.10100506696888
+        },
+        "candidate": {
+          "samples_ms": [
+            75.16223907470703,
+            75.09487915039062,
+            76.4610595703125,
+            75.77107238769531,
+            76.33715057373047,
+            76.33900451660156,
+            75.62342071533203,
+            76.34333038330078,
+            75.10733032226562,
+            75.13775634765625,
+            76.2429428100586,
+            75.65090942382812,
+            76.9865951538086,
+            76.45164489746094,
+            75.12268829345703,
+            76.36873626708984,
+            75.21382141113281,
+            75.16233825683594,
+            75.15340423583984,
+            75.63337707519531
+          ],
+          "p50_ms": 75.64214324951172,
+          "p95_ms": 76.48733634948731,
+          "mean_ms": 75.76818504333497,
+          "dense_tflop_s_p50": 77.04777466571349
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f",
+          "expected_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+        }
+      },
+      "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+    },
+    {
+      "selected_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "split_k": 1,
+        "reduction_scheme": 0,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "required_workspace_bytes": 0,
+        "waves": 1.0,
+        "state": 0,
+        "api_status": 0,
+        "valid": true,
+        "capabilities": {
+          "split_k_support": 1,
+          "reduction_scheme_mask": 6,
+          "cta_swizzle_support": 0,
+          "custom_option_max": 0,
+          "strided_batch_support": 1,
+          "out_of_place_result_support": 1,
+          "tile_ids": [
+            20
+          ],
+          "stages_ids": [
+            37
+          ],
+          "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+        }
+      },
+      "supplied_workspace_bytes": 67108864,
+      "required_workspace_bytes": 0,
+      "checked": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "split_k": 1,
+        "reduction_scheme": 0,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "required_workspace_bytes": 0,
+        "waves": 1.0,
+        "state": 0,
+        "api_status": 0,
+        "valid": true,
+        "capabilities": {
+          "split_k_support": 1,
+          "reduction_scheme_mask": 6,
+          "cta_swizzle_support": 0,
+          "custom_option_max": 0,
+          "strided_batch_support": 1,
+          "out_of_place_result_support": 1,
+          "tile_ids": [
+            20
+          ],
+          "stages_ids": [
+            37
+          ],
+          "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+        }
+      },
+      "fc2_only": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            53.98486328125,
+            53.7977294921875,
+            53.85007858276367,
+            53.532447814941406,
+            53.588897705078125,
+            53.95724868774414,
+            53.63395309448242,
+            53.011390686035156,
+            53.56748962402344,
+            54.085472106933594,
+            54.20851135253906,
+            53.808990478515625,
+            53.888832092285156,
+            54.48992156982422,
+            53.129215240478516,
+            53.52640151977539,
+            53.18348693847656,
+            53.60192108154297,
+            53.001953125,
+            53.25020980834961
+          ],
+          "p50_ms": 53.617937088012695,
+          "p95_ms": 54.22258186340332,
+          "mean_ms": 53.65495071411133,
+          "dense_tflop_s_p50": 108.69606562358724
+        },
+        "candidate": {
+          "samples_ms": [
+            15.689727783203125,
+            15.966943740844727,
+            15.744000434875488,
+            15.626208305358887,
+            15.685664176940918,
+            15.68841552734375,
+            15.331328392028809,
+            15.71504020690918,
+            14.792703628540039,
+            15.66592025756836,
+            14.791680335998535,
+            15.645536422729492,
+            14.812159538269043,
+            14.813887596130371,
+            14.793631553649902,
+            15.76848030090332,
+            15.557696342468262,
+            15.463264465332031,
+            14.7957763671875,
+            15.723199844360352
+          ],
+          "p50_ms": 15.63587236404419,
+          "p95_ms": 15.778403472900392,
+          "mean_ms": 15.403563261032104,
+          "dense_tflop_s_p50": 372.73640207770177
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f",
+          "expected_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+        }
+      },
+      "accepted_producer_plus_fc2": {
+        "order": "AB/BA alternates by round",
+        "baseline": {
+          "samples_ms": [
+            76.28396606445312,
+            76.32662200927734,
+            76.3351058959961,
+            76.94000244140625,
+            76.97782135009766,
+            76.59494018554688,
+            76.87884521484375,
+            77.03215789794922,
+            76.31446075439453,
+            76.24060821533203,
+            76.16102600097656,
+            76.39116668701172,
+            77.31603240966797,
+            76.89494323730469,
+            75.6058578491211,
+            76.90121459960938,
+            76.15897369384766,
+            77.05766296386719,
+            77.53421020507812,
+            76.94409942626953
+          ],
+          "p50_ms": 76.73689270019531,
+          "p95_ms": 77.32694129943847,
+          "mean_ms": 76.64448585510254,
+          "dense_tflop_s_p50": 75.94859008807853
+        },
+        "candidate": {
+          "samples_ms": [
+            38.138526916503906,
+            39.03964614868164,
+            39.00892639160156,
+            38.78895950317383,
+            38.25788879394531,
+            39.111358642578125,
+            37.374977111816406,
+            39.55583953857422,
+            38.676639556884766,
+            39.315616607666016,
+            39.011199951171875,
+            38.85724639892578,
+            38.27609634399414,
+            39.127777099609375,
+            38.3109130859375,
+            39.1360969543457,
+            38.23001480102539,
+            39.10319900512695,
+            38.683841705322266,
+            38.4785270690918
+          ],
+          "p50_ms": 38.823102951049805,
+          "p95_ms": 39.32762775421143,
+          "mean_ms": 38.72416458129883,
+          "dense_tflop_s_p50": 150.11831526368
+        },
+        "parity": {
+          "bf16_exact": true,
+          "different_elements": 0,
+          "max_abs": 0.0,
+          "mean_abs": 0.0,
+          "actual_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f",
+          "expected_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+        }
+      },
+      "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+    }
+  ],
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv b/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv
new file mode 100644
index 0000000..dd317f9
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv
@@ -0,0 +1,3 @@
+"ID","Process ID","Process Name","Host Name","Kernel Name","Context","Stream","Block Size","Grid Size","Device","CC","c2clink__enabled_mask","c2clink__present","derived__avg_thread_executed","derived__avg_thread_executed_true","derived__avg_thread_unexecuted_true","derived__derivative_avg_thread_executed_true","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.peak_sustained","derived__l1tex__lsu_writeback_bytes_mem_lgds.sum.per_second","derived__local_spilling_requests","derived__local_spilling_requests_pct","derived__lts__lts2xbar_bytes.sum.peak_sustained","derived__lts__lts2xbar_bytes.sum.per_second","derived__memory_l1_conflicts_shared_nway","derived__memory_l1_wavefronts_shared_excessive","derived__memory_l2_theoretical_sectors_global_excessive","derived__pct_occupancy_per_barrier_count","derived__pct_occupancy_per_block_size","derived__pct_occupancy_per_register_count","derived__pct_occupancy_per_shared_mem_size","derived__sm__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__sm__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__inst_executed_op_branch_pct","derived__smsp__sass_thread_inst_executed_op_dfma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_ffma_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hadd_pred_on_x2","derived__smsp__sass_thread_inst_executed_op_hfma_pred_on_x4","derived__smsp__sass_thread_inst_executed_op_hmul_pred_on_x2","derived_tempMetric0","derived_tempMetric1","derived_tempMetric2","derived_tempMetric3","derived_tempMetric4","derived_tempMetric5","device__attribute_architecture","device__attribute_async_engine_count","device__attribute_can_flush_remote_writes","device__attribute_can_map_host_memory","device__attribute_can_tex2d_gather","device__attribute_can_use_64_bit_stream_mem_ops","device__attribute_can_use_64_bit_stream_mem_ops_v1","device__attribute_can_use_host_pointer_for_registered_mem","device__attribute_can_use_stream_mem_ops_v1","device__attribute_can_use_stream_wait_value_nor","device__attribute_can_use_stream_wait_value_nor_v1","device__attribute_chip","device__attribute_clock_rate","device__attribute_cluster_launch","device__attribute_compute_capability_major","device__attribute_compute_capability_minor","device__attribute_compute_mode","device__attribute_compute_preemption_supported","device__attribute_concurrent_kernels","device__attribute_concurrent_managed_access","device__attribute_confidential_computing_mode","device__attribute_cooperative_launch","device__attribute_cooperative_multi_device_launch","device__attribute_deferred_mapping_cuda_array_supported","device__attribute_device_index","device__attribute_direct_managed_mem_access_from_host","device__attribute_display_name","device__attribute_dma_buf_supported","device__attribute_ecc_enabled","device__attribute_fb_bus_width","device__attribute_fbp_count","device__attribute_generic_compression_supported","device__attribute_global_l1_cache_supported","device__attribute_global_memory_bus_width","device__attribute_gpu_direct_rdma_flush_writes_options","device__attribute_gpu_direct_rdma_supported","device__attribute_gpu_direct_rdma_with_cuda_vmm_supported","device__attribute_gpu_direct_rdma_writes_ordering","device__attribute_gpu_overlap","device__attribute_gpu_pci_device_id","device__attribute_gpu_pci_ext_device_id","device__attribute_gpu_pci_ext_downstream_link_rate","device__attribute_gpu_pci_ext_downstream_link_width","device__attribute_gpu_pci_ext_gen","device__attribute_gpu_pci_ext_gpu_gen","device__attribute_gpu_pci_ext_gpu_link_rate","device__attribute_gpu_pci_ext_gpu_link_width","device__attribute_gpu_pci_revision_id","device__attribute_gpu_pci_sub_system_id","device__attribute_handle_type_fabric_supported","device__attribute_handle_type_posix_file_descriptor_supported","device__attribute_handle_type_win32_handle_supported","device__attribute_handle_type_win32_kmt_handle_supported","device__attribute_host_native_atomic_supported","device__attribute_host_numa_id","device__attribute_host_register_supported","device__attribute_implementation","device__attribute_integrated","device__attribute_ipc_event_supported","device__attribute_kernel_exec_timeout","device__attribute_l2_cache_size","device__attribute_l2s_count","device__attribute_limits_max_cta_per_sm","device__attribute_limits_num_tpcs","device__attribute_local_l1_cache_supported","device__attribute_managed_memory","device__attribute_max_access_policy_window_size","device__attribute_max_block_dim_x","device__attribute_max_block_dim_y","device__attribute_max_block_dim_z","device__attribute_max_blocks_per_multiprocessor","device__attribute_max_gpu_frequency_khz","device__attribute_max_grid_dim_x","device__attribute_max_grid_dim_y","device__attribute_max_grid_dim_z","device__attribute_max_ipc_per_multiprocessor","device__attribute_max_ipc_per_scheduler","device__attribute_max_mem_frequency_khz","device__attribute_max_persisting_l2_cache_size","device__attribute_max_pitch","device__attribute_max_registers_per_block","device__attribute_max_registers_per_multiprocessor","device__attribute_max_registers_per_thread","device__attribute_max_shared_memory_per_block","device__attribute_max_shared_memory_per_block_optin","device__attribute_max_shared_memory_per_multiprocessor","device__attribute_max_threads_per_block","device__attribute_max_threads_per_multiprocessor","device__attribute_max_warps_per_multiprocessor","device__attribute_max_warps_per_scheduler","device__attribute_maximum_surface1d_layered_layers","device__attribute_maximum_surface1d_layered_width","device__attribute_maximum_surface1d_width","device__attribute_maximum_surface2d_height","device__attribute_maximum_surface2d_layered_height","device__attribute_maximum_surface2d_layered_layers","device__attribute_maximum_surface2d_layered_width","device__attribute_maximum_surface2d_width","device__attribute_maximum_surface3d_depth","device__attribute_maximum_surface3d_height","device__attribute_maximum_surface3d_width","device__attribute_maximum_surfacecubemap_layered_layers","device__attribute_maximum_surfacecubemap_layered_width","device__attribute_maximum_surfacecubemap_width","device__attribute_maximum_texture1d_layered_layers","device__attribute_maximum_texture1d_layered_width","device__attribute_maximum_texture1d_linear_width","device__attribute_maximum_texture1d_mipmapped_width","device__attribute_maximum_texture1d_width","device__attribute_maximum_texture2d_gather_height","device__attribute_maximum_texture2d_gather_width","device__attribute_maximum_texture2d_height","device__attribute_maximum_texture2d_layered_height","device__attribute_maximum_texture2d_layered_layers","device__attribute_maximum_texture2d_layered_width","device__attribute_maximum_texture2d_linear_height","device__attribute_maximum_texture2d_linear_pitch","device__attribute_maximum_texture2d_linear_width","device__attribute_maximum_texture2d_mipmapped_height","device__attribute_maximum_texture2d_mipmapped_width","device__attribute_maximum_texture2d_width","device__attribute_maximum_texture3d_depth","device__attribute_maximum_texture3d_depth_alternate","device__attribute_maximum_texture3d_height","device__attribute_maximum_texture3d_height_alternate","device__attribute_maximum_texture3d_width","device__attribute_maximum_texture3d_width_alternate","device__attribute_maximum_texturecubemap_layered_layers","device__attribute_maximum_texturecubemap_layered_width","device__attribute_maximum_texturecubemap_width","device__attribute_mem_sync_domain_count","device__attribute_memory_clock_rate","device__attribute_memory_pools_supported","device__attribute_mempool_supported_handle_types","device__attribute_mps_enabled","device__attribute_multi_gpu_board","device__attribute_multi_gpu_board_group_id","device__attribute_multicast_supported","device__attribute_multiprocessor_count","device__attribute_num_l2s_per_fbp","device__attribute_num_schedulers_per_multiprocessor","device__attribute_num_tex_per_multiprocessor","device__attribute_numa_config","device__attribute_pageable_memory_access","device__attribute_pageable_memory_access_uses_host_page_tables","device__attribute_pci_bus_id","device__attribute_pci_device_id","device__attribute_pci_domain_id","device__attribute_ram_location","device__attribute_ram_type","device__attribute_reserved_shared_memory_per_block","device__attribute_sass_level","device__attribute_single_to_double_precision_perf_ratio","device__attribute_sparse_cuda_array_supported","device__attribute_stream_priorities_supported","device__attribute_surface_alignment","device__attribute_tcc_driver","device__attribute_tensor_map_access_supported","device__attribute_texture_alignment","device__attribute_texture_pitch_alignment","device__attribute_total_constant_memory","device__attribute_total_memory","device__attribute_unified_addressing","device__attribute_unified_function_pointers","device__attribute_virtual_address_management_supported","device__attribute_warp_size","gcc__cache_requests_type_constant.sum","gcc__cache_requests_type_constant.sum.pct_of_peak_sustained_elapsed","gcc__cache_requests_type_instruction.sum","gcc__cache_requests_type_instruction.sum.pct_of_peak_sustained_elapsed","gcc__xbar2gcc_sectors.sum","gcc__xbar2gcc_sectors.sum.pct_of_peak_sustained_elapsed","gpc__cycles_elapsed.avg","gpc__cycles_elapsed.avg.per_second","gpc__cycles_elapsed.max","gpc__cycles_elapsed.max.per_second","gpc__cycles_elapsed.min","gpc__cycles_elapsed.min.per_second","gpc__cycles_elapsed.sum","gpc__cycles_elapsed.sum.per_second","gpu__compute_memory_access_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_access_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_request_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.avg.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.max.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.min.pct_of_peak_sustained_elapsed","gpu__compute_memory_throughput.sum.pct_of_peak_sustained_elapsed","gpu__time_duration.avg","gpu__time_duration.max","gpu__time_duration.min","gpu__time_duration.sum","gr__workids_granted.avg","gr__workids_granted.max","gr__workids_granted.min","gr__workids_granted.sum","gr__workids_granted_as_ctas.avg","gr__workids_granted_as_ctas.max","gr__workids_granted_as_ctas.min","gr__workids_granted_as_ctas.sum","gr__workids_requested.avg","gr__workids_requested.max","gr__workids_requested.min","gr__workids_requested.sum","idc__request_cycles_active.avg.pct_of_peak_sustained_elapsed","idc__request_cycles_active.max.pct_of_peak_sustained_elapsed","idc__request_cycles_active.min.pct_of_peak_sustained_elapsed","idc__request_cycles_active.sum.pct_of_peak_sustained_elapsed","idc__request_hit_rate.pct","idc__requests.sum","idc__requests.sum.pct_of_peak_sustained_elapsed","inst_executed","l1tex__cycles_active.avg","l1tex__cycles_active.max","l1tex__cycles_active.min","l1tex__cycles_active.sum","l1tex__cycles_elapsed.avg","l1tex__cycles_elapsed.avg.per_second","l1tex__cycles_elapsed.max","l1tex__cycles_elapsed.max.per_second","l1tex__cycles_elapsed.min","l1tex__cycles_elapsed.min.per_second","l1tex__cycles_elapsed.sum","l1tex__cycles_elapsed.sum.per_second","l1tex__data_bank_conflicts_pipe_lsu_mem_shared.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_atom.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ld.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ldgsts.sum","l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_st.sum","l1tex__data_bank_reads.avg.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.max.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.min.pct_of_peak_sustained_elapsed","l1tex__data_bank_reads.sum.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.avg.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.max.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.min.pct_of_peak_sustained_elapsed","l1tex__data_bank_writes.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_atom.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum","l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__data_pipe_tex_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.avg.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.max.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.min.pct_of_peak_sustained_elapsed","l1tex__f_wavefronts.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.avg.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.max.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.min.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active_mem_lgds.sum","l1tex__lsu_writeback_active_mem_lgds.sum.pct_of_peak_sustained_elapsed","l1tex__lsu_writeback_active_mem_lgds.sum.peak_sustained","l1tex__lsu_writeback_active_mem_lgds.sum.per_second","l1tex__lsuin_requests.avg.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.max.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.min.pct_of_peak_sustained_elapsed","l1tex__lsuin_requests.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.avg.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.max.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.min.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_req_cycles_active.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes.sum","l1tex__m_l1tex2xbar_write_bytes.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_dshared.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_red.sum.per_second","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_bytes_mem_global_op_tma_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_atom.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_ld.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_redas.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_dshared_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_red.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_global_op_tma_st.sum.per_second","l1tex__m_l1tex2xbar_write_sectors_mem_lg_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_lg_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_atom.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_red.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_st.sum","l1tex__m_l1tex2xbar_write_sectors_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes.sum","l1tex__m_xbar2l1tex_read_bytes.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes.sum.per_second","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes_mem_dshared.sum.per_second","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_bytes_mem_global_op_tma_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors.avg.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.max.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.min.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_atom.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_redas.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_st.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_red.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_red.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_st.sum","l1tex__m_xbar2l1tex_read_sectors_mem_dshared_op_tma_st.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_global_op_tma_ld.sum.per_second","l1tex__m_xbar2l1tex_read_sectors_mem_lg_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_lg_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_atom.sum","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_ld.sum","l1tex__m_xbar2l1tex_read_sectors_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__m_xbar2l1tex_read_sectors_mem_texture.sum","l1tex__m_xbar2l1tex_read_sectors_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","l1tex__t_bytes_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.per_second","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_atom.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_red.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_global_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_ld.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_st.sum","l1tex__t_output_wavefronts_pipe_lsu_mem_local_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_atom.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_ld.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_red.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_st.sum","l1tex__t_output_wavefronts_pipe_tex_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_output_wavefronts_pipe_tex_mem_texture.sum","l1tex__t_output_wavefronts_pipe_tex_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_redas.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_requests_pipe_lsu_mem_dshared_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum","l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_global_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_red.sum","l1tex__t_requests_pipe_lsu_mem_global_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_global_op_st.sum","l1tex__t_requests_pipe_lsu_mem_global_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_local_op_ld.sum","l1tex__t_requests_pipe_lsu_mem_local_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_lsu_mem_local_op_st.sum","l1tex__t_requests_pipe_lsu_mem_local_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_atom.sum","l1tex__t_requests_pipe_tex_mem_surface_op_atom.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_ld.sum","l1tex__t_requests_pipe_tex_mem_surface_op_ld.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_red.sum","l1tex__t_requests_pipe_tex_mem_surface_op_red.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_surface_op_st.sum","l1tex__t_requests_pipe_tex_mem_surface_op_st.sum.pct_of_peak_sustained_elapsed","l1tex__t_requests_pipe_tex_mem_texture.sum","l1tex__t_requests_pipe_tex_mem_texture.sum.pct_of_peak_sustained_elapsed","l1tex__t_sector_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_atom_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_ld_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_red_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_global_op_st_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_local_op_ld_hit_rate.pct","l1tex__t_sector_pipe_lsu_mem_local_op_st_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_atom_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_ld_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_red_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_surface_op_st_hit_rate.pct","l1tex__t_sector_pipe_tex_mem_texture_op_tex_hit_rate.pct","l1tex__t_sectors_pipe_lsu_mem_dshared_op_atom.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_redas.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_redas_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_dshared_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_red_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_global_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_lsu_mem_local_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_atom_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_ld_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_red_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_surface_op_st_lookup_miss.sum","l1tex__t_sectors_pipe_tex_mem_texture.sum","l1tex__t_sectors_pipe_tex_mem_texture_lookup_hit.sum","l1tex__t_sectors_pipe_tex_mem_texture_lookup_miss.sum","l1tex__tex_writeback_active.avg.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.max.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.min.pct_of_peak_sustained_elapsed","l1tex__tex_writeback_active.sum","l1tex__tex_writeback_active.sum.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.avg.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.max.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.min.pct_of_peak_sustained_elapsed","l1tex__texin_sm2tex_req_cycles_active.sum.pct_of_peak_sustained_elapsed","l1tex__throughput.avg.pct_of_peak_sustained_active","l1tex__throughput.avg.pct_of_peak_sustained_elapsed","l1tex__throughput.max.pct_of_peak_sustained_active","l1tex__throughput.max.pct_of_peak_sustained_elapsed","l1tex__throughput.min.pct_of_peak_sustained_active","l1tex__throughput.min.pct_of_peak_sustained_elapsed","l1tex__throughput.sum.pct_of_peak_sustained_active","l1tex__throughput.sum.pct_of_peak_sustained_elapsed","launch__barrier_count","launch__block_dim_x","launch__block_dim_y","launch__block_dim_z","launch__block_size","launch__cluster_dim_x","launch__cluster_dim_y","launch__cluster_dim_z","launch__cluster_max_active","launch__cluster_max_potential_size","launch__cluster_scheduling_policy","launch__cluster_size","launch__context_id","launch__device_id","launch__func_cache_config","launch__function_pcs","launch__grid_dim_x","launch__grid_dim_y","launch__grid_dim_z","launch__grid_size","launch__kernel_name","launch__occupancy_cluster_gpu_pct","launch__occupancy_cluster_pct","launch__occupancy_limit_barriers","launch__occupancy_limit_blocks","launch__occupancy_limit_registers","launch__occupancy_limit_shared_mem","launch__occupancy_limit_warps","launch__occupancy_per_barrier_count","launch__occupancy_per_block_size","launch__occupancy_per_cluster_size","launch__occupancy_per_register_count","launch__occupancy_per_shared_mem_size","launch__persisting_l2_cache_size","launch__preferred_cluster_dim_x","launch__preferred_cluster_dim_y","launch__preferred_cluster_dim_z","launch__preferred_cluster_size","launch__registers_per_thread","launch__registers_per_thread_allocated","launch__shared_mem_config_size","launch__shared_mem_per_block","launch__shared_mem_per_block_allocated","launch__shared_mem_per_block_driver","launch__shared_mem_per_block_dynamic","launch__shared_mem_per_block_static","launch__sm_count","launch__stack_size","launch__stream_id","launch__thread_count","launch__tpc_count","launch__tpc_enabled","launch__uses_cdp","launch__uses_green_context","launch__uses_mps","launch__uses_nvlink_centric_scheduling","launch__uses_vgpu","launch__waves_per_multiprocessor","lrc__average_ilc_input_sector_success_rate.pct","lrc__ilc_input_sectors.sum","lts__average_gcomp_input_sector_success_rate.pct","lts__average_gcomp_output_sector_compression_achieved_rate.ratio","lts__average_t_sectors_requested_srcunit_tex_op_read_vs_returned_sectors_realtime.ratio","lts__average_xcomp_gxc_sector_input_compression_rate.ratio","lts__cycles_active.avg","lts__cycles_active.max","lts__cycles_active.min","lts__cycles_active.sum","lts__cycles_elapsed.avg","lts__cycles_elapsed.avg.per_second","lts__cycles_elapsed.max","lts__cycles_elapsed.max.per_second","lts__cycles_elapsed.min","lts__cycles_elapsed.min.per_second","lts__cycles_elapsed.sum","lts__cycles_elapsed.sum.per_second","lts__d_atomic_input_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.max.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.min.pct_of_peak_sustained_elapsed","lts__d_atomic_input_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.avg.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.max.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.min.pct_of_peak_sustained_elapsed","lts__d_decomp_input_sectors.sum.pct_of_peak_sustained_elapsed","lts__d_sectors.avg.pct_of_peak_sustained_elapsed","lts__d_sectors.max.pct_of_peak_sustained_elapsed","lts__d_sectors.min.pct_of_peak_sustained_elapsed","lts__d_sectors.sum.pct_of_peak_sustained_elapsed","lts__d_sectors_fill_sysmem.sum","lts__d_sectors_fill_sysmem.sum.pct_of_peak_sustained_elapsed","lts__d_sectors_fill_sysmem.sum.per_second","lts__gcomp_input_sectors.avg","lts__gcomp_input_sectors.max","lts__gcomp_input_sectors.min","lts__gcomp_input_sectors.sum","lts__lts2xbar_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.avg.peak_sustained","lts__lts2xbar_cycles_active.avg.per_second","lts__lts2xbar_cycles_active.max.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.max.peak_sustained","lts__lts2xbar_cycles_active.max.per_second","lts__lts2xbar_cycles_active.min.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.min.peak_sustained","lts__lts2xbar_cycles_active.min.per_second","lts__lts2xbar_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__lts2xbar_cycles_active.sum.peak_sustained","lts__lts2xbar_cycles_active.sum.per_second","lts__t_requests.sum","lts__t_requests_srcunit_gcc.sum","lts__t_requests_srcunit_tex.sum","lts__t_requests_srcunit_tex_op_atom_dot_alu.sum","lts__t_requests_srcunit_tex_op_atom_dot_cas.sum","lts__t_requests_srcunit_tex_op_read.sum","lts__t_requests_srcunit_tex_op_red.sum","lts__t_requests_srcunit_tex_op_write.sum","lts__t_sector_hit_rate.pct","lts__t_sector_op_read_hit_rate.pct","lts__t_sector_op_write_hit_rate.pct","lts__t_sectors.avg","lts__t_sectors.avg.pct_of_peak_sustained_elapsed","lts__t_sectors.avg.peak_sustained","lts__t_sectors.avg.per_cycle_elapsed","lts__t_sectors.max","lts__t_sectors.max.pct_of_peak_sustained_elapsed","lts__t_sectors.min","lts__t_sectors.min.pct_of_peak_sustained_elapsed","lts__t_sectors.sum","lts__t_sectors.sum.pct_of_peak_sustained_elapsed","lts__t_sectors.sum.per_second","lts__t_sectors_aperture_sysmem_op_write.sum","lts__t_sectors_aperture_sysmem_op_write.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_aperture_sysmem_op_write.sum.per_second","lts__t_sectors_data_ecc.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_data_ecc.avg.peak_sustained","lts__t_sectors_data_ecc.avg.per_cycle_elapsed","lts__t_sectors_data_ecc.sum","lts__t_sectors_data_ecc.sum.per_second","lts__t_sectors_evict_first_lookup_hit.sum","lts__t_sectors_evict_first_lookup_miss.sum","lts__t_sectors_evict_last_lookup_hit.sum","lts__t_sectors_evict_last_lookup_miss.sum","lts__t_sectors_evict_normal_demote_lookup_hit.sum","lts__t_sectors_evict_normal_demote_lookup_miss.sum","lts__t_sectors_evict_normal_lookup_hit.sum","lts__t_sectors_evict_normal_lookup_miss.sum","lts__t_sectors_lookup_hit.sum","lts__t_sectors_lookup_miss.sum","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_requested_srcunit_tex_op_read_realtime.sum.per_second","lts__t_sectors_srcunit_gcc.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_gcc.avg.peak_sustained","lts__t_sectors_srcunit_gcc.avg.per_cycle_elapsed","lts__t_sectors_srcunit_gcc.sum","lts__t_sectors_srcunit_gcc.sum.per_second","lts__t_sectors_srcunit_gcc_lookup_hit.sum","lts__t_sectors_srcunit_gcc_lookup_miss.sum","lts__t_sectors_srcunit_tex.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.avg.peak_sustained","lts__t_sectors_srcunit_tex.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex.max.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.min.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.sum","lts__t_sectors_srcunit_tex.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex.sum.per_second","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.avg","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.max","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.min","lts__t_sectors_srcunit_tex_aperture_peer_lookup_miss.sum","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.avg","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.max","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.min","lts__t_sectors_srcunit_tex_aperture_sysmem_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_lookup_hit.sum","lts__t_sectors_srcunit_tex_lookup_miss.avg","lts__t_sectors_srcunit_tex_lookup_miss.max","lts__t_sectors_srcunit_tex_lookup_miss.min","lts__t_sectors_srcunit_tex_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom.sum","lts__t_sectors_srcunit_tex_op_atom.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_atom_dot_alu.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_alu.sum","lts__t_sectors_srcunit_tex_op_atom_dot_alu.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_alu_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_dot_alu_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_atom_dot_cas.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_atom_dot_cas.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas.sum.per_second","lts__t_sectors_srcunit_tex_op_atom_dot_cas_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_dot_cas_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_atom_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_atom_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_read.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_read.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_read.sum","lts__t_sectors_srcunit_tex_op_read.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_read.sum.per_second","lts__t_sectors_srcunit_tex_op_read_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_read_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_read_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_red.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_red.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_red.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_red.sum","lts__t_sectors_srcunit_tex_op_red.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_red.sum.per_second","lts__t_sectors_srcunit_tex_op_red_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_red_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write.avg.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_write.avg.peak_sustained","lts__t_sectors_srcunit_tex_op_write.avg.per_cycle_elapsed","lts__t_sectors_srcunit_tex_op_write.sum","lts__t_sectors_srcunit_tex_op_write.sum.pct_of_peak_sustained_elapsed","lts__t_sectors_srcunit_tex_op_write.sum.per_second","lts__t_sectors_srcunit_tex_op_write_evict_first_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_first_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_last_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_last_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_demote_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_demote_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_evict_normal_lookup_miss.sum","lts__t_sectors_srcunit_tex_op_write_lookup_hit.sum","lts__t_sectors_srcunit_tex_op_write_lookup_miss.sum","lts__t_tag_requests.avg.pct_of_peak_sustained_elapsed","lts__t_tag_requests.max.pct_of_peak_sustained_elapsed","lts__t_tag_requests.min.pct_of_peak_sustained_elapsed","lts__t_tag_requests.sum.pct_of_peak_sustained_elapsed","lts__throughput.avg.pct_of_peak_sustained_elapsed","lts__throughput.max.pct_of_peak_sustained_elapsed","lts__throughput.min.pct_of_peak_sustained_elapsed","lts__throughput.sum.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.avg.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.max.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.min.pct_of_peak_sustained_elapsed","lts__xbar2lts_cycles_active.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_input.sum","lts__xcomp_gxc_sectors_input.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_input.sum.per_second","lts__xcomp_gxc_sectors_output.sum","lts__xcomp_gxc_sectors_output.sum.pct_of_peak_sustained_elapsed","lts__xcomp_gxc_sectors_output.sum.per_second","memory_access_size_type","memory_access_type","memory_l1_tag_requests_global","memory_l1_wavefronts_shared","memory_l1_wavefronts_shared_ideal","memory_l2_theoretical_sectors_global","memory_l2_theoretical_sectors_global_ideal","memory_l2_theoretical_sectors_local","memory_type","numa__cpu_affinity","numa__dev_display_name_all","numa__id_cpu","numa__id_memory","nvlink__bandwidth","nvlink__count_logical","nvlink__count_physical","nvlink__destination_ports","nvlink__dev0Id","nvlink__dev0type","nvlink__dev1Id","nvlink__dev1type","nvlink__dev_display_name_all","nvlink__enabled_mask","nvlink__is_direct_link","nvlink__is_nvswitch_connected","nvlink__max_count","nvlink__peer_access","nvlink__peer_atomic","nvlink__source_ports","nvlink__system_access","nvlink__system_atomic","pmsampling:gr__ctas_launched_queue_sync_realtime.sum","pmsampling:l1tex__data_pipe_lsu_wavefronts.avg","pmsampling:l1tex__lsu_writeback_active.avg","pmsampling:l1tex__t_sector_hit_rate.pct","pmsampling:lts__average_t_sector_hit_rate_realtime.pct","pmsampling:sm__average_thread_inst_executed_pred_on_per_inst_executed_realtime.pct","pmsampling:sm__cycles_active.avg","pmsampling:sm__inst_executed_pipe_alu_realtime.avg.pct_of_peak_sustained_elapsed","pmsampling:sm__inst_executed_realtime.avg.per_cycle_active","pmsampling:sm__pipe_tensor_cycles_active_realtime.avg.pct_of_peak_sustained_elapsed","pmsampling:smsp__inst_executed_pipe_fmaheavy.avg.pct_of_peak_sustained_elapsed","pmsampling:smsp__inst_executed_pipe_fmalite.avg.pct_of_peak_sustained_elapsed","pmsampling:tpc__warps_active_realtime.avg.per_cycle_active","pmsampling:tpc__warps_active_realtime.sum.per_cycle_active","profiler__perfworks_session_reuse","profiler__pmsampler_buffer_size_bytes","profiler__pmsampler_ctxsw_0","profiler__pmsampler_ctxsw_1","profiler__pmsampler_interval_time","profiler__pmsampler_merged_samples","profiler__pmsampler_pass_groups","profiler__replayer_bytes_mem_accessible.avg","profiler__replayer_bytes_mem_accessible.max","profiler__replayer_bytes_mem_accessible.min","profiler__replayer_bytes_mem_accessible.sum","profiler__replayer_bytes_mem_backed_up.avg","profiler__replayer_bytes_mem_backed_up.max","profiler__replayer_bytes_mem_backed_up.min","profiler__replayer_bytes_mem_backed_up.sum","profiler__replayer_passes","profiler__replayer_passes_type_warmup","profiler__timestamp_workload_end_0","profiler__timestamp_workload_end_1","profiler__timestamp_workload_start_0","profiler__timestamp_workload_start_1","sass__inst_executed_global_loads","sass__inst_executed_global_stores","sass__inst_executed_local_loads","sass__inst_executed_local_stores","sass__inst_executed_per_opcode","sass__inst_executed_per_opcode_category","sass__inst_executed_per_opcode_with_modifier_all","sass__inst_executed_per_opcode_with_modifier_selective","sass__inst_executed_register_spilling","sass__inst_executed_register_spilling_mem_local","sass__inst_executed_register_spilling_mem_shared","sass__inst_executed_register_spilling_op_read","sass__inst_executed_register_spilling_op_write","sass__inst_executed_shared_loads","sass__inst_executed_shared_stores","sass__thread_inst_executed_per_opcode_category","sass__thread_inst_executed_true_per_opcode","sass__thread_inst_executed_true_per_opcode_with_modifier_all","sass__thread_inst_executed_true_per_opcode_with_modifier_selective","sm__cycles_active.avg","sm__cycles_active.max","sm__cycles_active.min","sm__cycles_active.sum","sm__cycles_elapsed.avg","sm__cycles_elapsed.avg.per_second","sm__cycles_elapsed.max","sm__cycles_elapsed.max.per_second","sm__cycles_elapsed.min","sm__cycles_elapsed.min.per_second","sm__cycles_elapsed.sum","sm__cycles_elapsed.sum.per_second","sm__dcc_request_hit_rate.pct","sm__dcc_requests.sum","sm__dcc_requests.sum.pct_of_peak_sustained_elapsed","sm__dcc_requests_lookup_hit.sum","sm__icc_request_hit_rate.pct","sm__icc_requests.sum","sm__icc_requests.sum.pct_of_peak_sustained_elapsed","sm__inst_executed.avg.pct_of_peak_sustained_elapsed","sm__inst_executed.avg.per_cycle_active","sm__inst_executed.avg.per_cycle_elapsed","sm__inst_executed.max.pct_of_peak_sustained_elapsed","sm__inst_executed.max.per_cycle_active","sm__inst_executed.max.per_cycle_elapsed","sm__inst_executed.min.pct_of_peak_sustained_elapsed","sm__inst_executed.min.per_cycle_active","sm__inst_executed.min.per_cycle_elapsed","sm__inst_executed.sum.pct_of_peak_sustained_elapsed","sm__inst_executed.sum.per_cycle_active","sm__inst_executed.sum.per_cycle_elapsed","sm__inst_executed_pipe_adu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_adu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_adu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_alu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_alu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_cbu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_cbu_pred_on_any.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fma_type_fp16.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fma_type_fp16.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_dmma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_dmma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_fp64_op_fp64.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_fp64_op_fp64.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_ipa.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_lsu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_lsu.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tensor_subpipe_hmma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tensor_subpipe_hmma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tensor_subpipe_imma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tensor_subpipe_imma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tex.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_tex.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_tma.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_tma.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_uniform.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_uniform.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_workid.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_workid.sum.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.avg.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.avg.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.max.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.max.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.min.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.min.pct_of_peak_sustained_elapsed","sm__inst_executed_pipe_xu.sum.pct_of_peak_sustained_active","sm__inst_executed_pipe_xu.sum.pct_of_peak_sustained_elapsed","sm__inst_issued.avg.pct_of_peak_sustained_elapsed","sm__inst_issued.avg.per_cycle_active","sm__inst_issued.max.pct_of_peak_sustained_elapsed","sm__inst_issued.max.per_cycle_active","sm__inst_issued.min.pct_of_peak_sustained_elapsed","sm__inst_issued.min.per_cycle_active","sm__inst_issued.sum.pct_of_peak_sustained_elapsed","sm__inst_issued.sum.per_cycle_active","sm__instruction_throughput.avg.pct_of_peak_sustained_elapsed","sm__instruction_throughput.max.pct_of_peak_sustained_elapsed","sm__instruction_throughput.min.pct_of_peak_sustained_elapsed","sm__instruction_throughput.sum.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","sm__instruction_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","sm__issue_active.avg.pct_of_peak_sustained_elapsed","sm__issue_active.max.pct_of_peak_sustained_elapsed","sm__issue_active.min.pct_of_peak_sustained_elapsed","sm__issue_active.sum.pct_of_peak_sustained_elapsed","sm__maximum_warps_avg_per_active_cycle","sm__maximum_warps_per_active_cycle_pct","sm__memory_throughput.avg.pct_of_peak_sustained_elapsed","sm__memory_throughput.max.pct_of_peak_sustained_elapsed","sm__memory_throughput.min.pct_of_peak_sustained_elapsed","sm__memory_throughput.sum.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.avg.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.max.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.min.pct_of_peak_sustained_elapsed","sm__memory_throughput_internal_activity.sum.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.avg.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.max.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.min.pct_of_peak_sustained_elapsed","sm__mio2rf_writeback_active.sum.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.avg.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.max.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.min.pct_of_peak_sustained_elapsed","sm__mio_inst_issued.sum.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.max.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.min.pct_of_peak_sustained_elapsed","sm__mio_pq_read_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.max.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.min.pct_of_peak_sustained_elapsed","sm__mio_pq_write_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_bf16_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_fp16_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_hmma_src_tf32_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.avg.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.max.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.min.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_off.sum.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.avg.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.max.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.min.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_op_imma_src_int8_sparsity_on.sum.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.avg","sm__ops_path_tensor_src_bf16_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.avg.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.max","sm__ops_path_tensor_src_bf16_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.max.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.min","sm__ops_path_tensor_src_bf16_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.min.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.sum","sm__ops_path_tensor_src_bf16_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_bf16_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_bf16_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_bf16_dst_fp32.sum.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.avg","sm__ops_path_tensor_src_fp16_dst_fp16.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.avg.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.avg.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.max","sm__ops_path_tensor_src_fp16_dst_fp16.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.max.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.max.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.min","sm__ops_path_tensor_src_fp16_dst_fp16.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.min.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.min.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.sum","sm__ops_path_tensor_src_fp16_dst_fp16.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.sum.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp16.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp16.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp16.sum.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.avg","sm__ops_path_tensor_src_fp16_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.avg.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.max","sm__ops_path_tensor_src_fp16_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.max.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.min","sm__ops_path_tensor_src_fp16_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.min.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.sum","sm__ops_path_tensor_src_fp16_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_fp16_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp16_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp16_dst_fp32.sum.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp4_fp6_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp64.avg","sm__ops_path_tensor_src_fp64.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.avg.peak_sustained","sm__ops_path_tensor_src_fp64.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.avg.per_second","sm__ops_path_tensor_src_fp64.max","sm__ops_path_tensor_src_fp64.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.max.peak_sustained","sm__ops_path_tensor_src_fp64.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.max.per_second","sm__ops_path_tensor_src_fp64.min","sm__ops_path_tensor_src_fp64.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.min.peak_sustained","sm__ops_path_tensor_src_fp64.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.min.per_second","sm__ops_path_tensor_src_fp64.sum","sm__ops_path_tensor_src_fp64.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp64.sum.peak_sustained","sm__ops_path_tensor_src_fp64.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp64.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp64.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp16_sparsity_on.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_off.sum.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.avg.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.max.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.min.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.peak_sustained","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.per_cycle_elapsed","sm__ops_path_tensor_src_fp8_dst_fp32_sparsity_on.sum.per_second","sm__ops_path_tensor_src_int8.avg","sm__ops_path_tensor_src_int8.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.avg.peak_sustained","sm__ops_path_tensor_src_int8.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.avg.per_cycle_elapsed","sm__ops_path_tensor_src_int8.avg.per_second","sm__ops_path_tensor_src_int8.max","sm__ops_path_tensor_src_int8.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.max.peak_sustained","sm__ops_path_tensor_src_int8.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.max.per_cycle_elapsed","sm__ops_path_tensor_src_int8.max.per_second","sm__ops_path_tensor_src_int8.min","sm__ops_path_tensor_src_int8.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.min.peak_sustained","sm__ops_path_tensor_src_int8.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.min.per_cycle_elapsed","sm__ops_path_tensor_src_int8.min.per_second","sm__ops_path_tensor_src_int8.sum","sm__ops_path_tensor_src_int8.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_int8.sum.peak_sustained","sm__ops_path_tensor_src_int8.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_int8.sum.per_cycle_elapsed","sm__ops_path_tensor_src_int8.sum.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.avg","sm__ops_path_tensor_src_tf32_dst_fp32.avg.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.avg.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.avg.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.avg.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.avg.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.max","sm__ops_path_tensor_src_tf32_dst_fp32.max.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.max.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.max.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.max.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.max.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.min","sm__ops_path_tensor_src_tf32_dst_fp32.min.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.min.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.min.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.min.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.min.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.sum","sm__ops_path_tensor_src_tf32_dst_fp32.sum.pct_of_peak_sustained_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.sum.peak_sustained","sm__ops_path_tensor_src_tf32_dst_fp32.sum.peak_sustained_elapsed.per_second","sm__ops_path_tensor_src_tf32_dst_fp32.sum.per_cycle_elapsed","sm__ops_path_tensor_src_tf32_dst_fp32.sum.per_second","sm__pipe_alu_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_alu_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_alu_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_aluheavy_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fma_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_fma_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fmaheavy_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_fp64_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_fp64_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_tensor_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_tensor_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__pipe_tensor_subpipe_hmma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_subpipe_hmma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tensor_subpipe_imma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tensor_subpipe_imma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.avg.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.avg.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.max.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.max.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.min.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.min.pct_of_peak_sustained_elapsed","sm__pipe_tma_cycles_active.sum.pct_of_peak_sustained_active","sm__pipe_tma_cycles_active.sum.pct_of_peak_sustained_elapsed","sm__sass_inst_executed_op_ldgsts_cache_access.sum","sm__sass_inst_executed_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_m_xbar2l1tex_read_bytes_mem_global_op_ldgsts_cache_bypass.sum.per_second","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts.sum","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","sm__sass_l1tex_t_requests_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_access.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum","sm__sass_l1tex_t_sectors_pipe_lsu_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","sm__sass_thread_inst_executed_op_dfma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_dfma_pred_on.sum.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_ffma_pred_on.sum.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.avg.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.max.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.min.peak_sustained","sm__sass_thread_inst_executed_op_hfma_pred_on.sum.peak_sustained","sm__throughput.avg.pct_of_peak_sustained_elapsed","sm__throughput.max.pct_of_peak_sustained_elapsed","sm__throughput.min.pct_of_peak_sustained_elapsed","sm__throughput.sum.pct_of_peak_sustained_elapsed","sm__warps_active.avg.pct_of_peak_sustained_active","sm__warps_active.avg.per_cycle_active","sm__warps_active.max.pct_of_peak_sustained_active","sm__warps_active.max.per_cycle_active","sm__warps_active.min.pct_of_peak_sustained_active","sm__warps_active.min.per_cycle_active","sm__warps_active.sum.pct_of_peak_sustained_active","sm__warps_active.sum.per_cycle_active","smsp__average_warp_latency_per_inst_issued.ratio","smsp__average_warps_active_per_inst_executed.ratio","smsp__average_warps_issue_stalled_barrier_per_issue_active.ratio","smsp__average_warps_issue_stalled_branch_resolving_per_issue_active.ratio","smsp__average_warps_issue_stalled_dispatch_stall_per_issue_active.ratio","smsp__average_warps_issue_stalled_drain_per_issue_active.ratio","smsp__average_warps_issue_stalled_lg_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_long_scoreboard_per_issue_active.ratio","smsp__average_warps_issue_stalled_math_pipe_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_membar_per_issue_active.ratio","smsp__average_warps_issue_stalled_mio_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_misc_per_issue_active.ratio","smsp__average_warps_issue_stalled_no_instruction_per_issue_active.ratio","smsp__average_warps_issue_stalled_not_selected_per_issue_active.ratio","smsp__average_warps_issue_stalled_selected_per_issue_active.ratio","smsp__average_warps_issue_stalled_short_scoreboard_per_issue_active.ratio","smsp__average_warps_issue_stalled_sleeping_per_issue_active.ratio","smsp__average_warps_issue_stalled_tex_throttle_per_issue_active.ratio","smsp__average_warps_issue_stalled_wait_per_issue_active.ratio","smsp__branch_targets_threads_divergent","smsp__cycles_active.avg","smsp__cycles_active.max","smsp__cycles_active.min","smsp__cycles_active.sum","smsp__cycles_elapsed.avg","smsp__cycles_elapsed.avg.per_second","smsp__cycles_elapsed.max","smsp__cycles_elapsed.max.per_second","smsp__cycles_elapsed.min","smsp__cycles_elapsed.min.per_second","smsp__cycles_elapsed.sum","smsp__cycles_elapsed.sum.per_second","smsp__inst_executed.avg","smsp__inst_executed.max","smsp__inst_executed.min","smsp__inst_executed.sum","smsp__inst_executed_op_branch.avg","smsp__inst_executed_op_branch.max","smsp__inst_executed_op_branch.min","smsp__inst_executed_op_branch.sum","smsp__inst_executed_op_generic_atom_dot_alu.sum","smsp__inst_executed_op_generic_atom_dot_alu.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_generic_atom_dot_cas.sum","smsp__inst_executed_op_generic_atom_dot_cas.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_global_red.sum","smsp__inst_executed_op_global_red.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_ldgsts.sum","smsp__inst_executed_op_ldgsts.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_ldsm.sum","smsp__inst_executed_op_ldsm.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_shared_atom.sum","smsp__inst_executed_op_shared_atom.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_shared_stsm.sum","smsp__inst_executed_op_shared_stsm.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_atom_dot_alu.sum","smsp__inst_executed_op_surface_atom_dot_alu.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_atom_dot_cas.sum","smsp__inst_executed_op_surface_atom_dot_cas.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_ld.sum","smsp__inst_executed_op_surface_ld.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_red.sum","smsp__inst_executed_op_surface_red.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_surface_st.sum","smsp__inst_executed_op_surface_st.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_texture.sum","smsp__inst_executed_op_texture.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_tma_ld.sum","smsp__inst_executed_op_tma_ld.sum.pct_of_peak_sustained_elapsed","smsp__inst_executed_op_tma_st.sum","smsp__inst_executed_op_tma_st.sum.pct_of_peak_sustained_elapsed","smsp__inst_issued.avg","smsp__inst_issued.max","smsp__inst_issued.min","smsp__inst_issued.sum","smsp__issue_active.avg.pct_of_peak_sustained_active","smsp__issue_active.avg.per_cycle_active","smsp__issue_active.max.pct_of_peak_sustained_active","smsp__issue_active.max.per_cycle_active","smsp__issue_active.min.pct_of_peak_sustained_active","smsp__issue_active.min.per_cycle_active","smsp__issue_active.sum.pct_of_peak_sustained_active","smsp__issue_active.sum.per_cycle_active","smsp__issue_inst0.avg.pct_of_peak_sustained_active","smsp__issue_inst0.max.pct_of_peak_sustained_active","smsp__issue_inst0.min.pct_of_peak_sustained_active","smsp__issue_inst0.sum.pct_of_peak_sustained_active","smsp__maximum_warps_avg_per_active_cycle","smsp__pcsamp_aggregated_passes","smsp__pcsamp_buffer_size_bytes","smsp__pcsamp_dropped_bytes","smsp__pcsamp_interval","smsp__pcsamp_interval_cycles","smsp__pcsamp_sample_count","smsp__pcsamp_warps_issue_stalled_barrier","smsp__pcsamp_warps_issue_stalled_barrier_not_issued","smsp__pcsamp_warps_issue_stalled_branch_resolving","smsp__pcsamp_warps_issue_stalled_branch_resolving_not_issued","smsp__pcsamp_warps_issue_stalled_dispatch_stall","smsp__pcsamp_warps_issue_stalled_dispatch_stall_not_issued","smsp__pcsamp_warps_issue_stalled_drain","smsp__pcsamp_warps_issue_stalled_drain_not_issued","smsp__pcsamp_warps_issue_stalled_lg_throttle","smsp__pcsamp_warps_issue_stalled_lg_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_long_scoreboard","smsp__pcsamp_warps_issue_stalled_long_scoreboard_not_issued","smsp__pcsamp_warps_issue_stalled_math_pipe_throttle","smsp__pcsamp_warps_issue_stalled_math_pipe_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_membar","smsp__pcsamp_warps_issue_stalled_membar_not_issued","smsp__pcsamp_warps_issue_stalled_mio_throttle","smsp__pcsamp_warps_issue_stalled_mio_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_misc","smsp__pcsamp_warps_issue_stalled_misc_not_issued","smsp__pcsamp_warps_issue_stalled_no_instructions","smsp__pcsamp_warps_issue_stalled_no_instructions_not_issued","smsp__pcsamp_warps_issue_stalled_not_selected","smsp__pcsamp_warps_issue_stalled_not_selected_not_issued","smsp__pcsamp_warps_issue_stalled_selected","smsp__pcsamp_warps_issue_stalled_selected_not_issued","smsp__pcsamp_warps_issue_stalled_short_scoreboard","smsp__pcsamp_warps_issue_stalled_short_scoreboard_not_issued","smsp__pcsamp_warps_issue_stalled_sleeping","smsp__pcsamp_warps_issue_stalled_sleeping_not_issued","smsp__pcsamp_warps_issue_stalled_tex_throttle","smsp__pcsamp_warps_issue_stalled_tex_throttle_not_issued","smsp__pcsamp_warps_issue_stalled_wait","smsp__pcsamp_warps_issue_stalled_wait_not_issued","smsp__sass_average_branch_targets_threads_uniform.pct","smsp__sass_average_data_bytes_per_sector_mem_global_op_ld.max_rate","smsp__sass_average_data_bytes_per_sector_mem_global_op_ld.ratio","smsp__sass_average_data_bytes_per_sector_mem_global_op_st.max_rate","smsp__sass_average_data_bytes_per_sector_mem_global_op_st.ratio","smsp__sass_average_data_bytes_per_sector_mem_local_op_ld.max_rate","smsp__sass_average_data_bytes_per_sector_mem_local_op_ld.ratio","smsp__sass_average_data_bytes_per_sector_mem_local_op_st.max_rate","smsp__sass_average_data_bytes_per_sector_mem_local_op_st.ratio","smsp__sass_branch_targets_threads_divergent.avg","smsp__sass_branch_targets_threads_divergent.max","smsp__sass_branch_targets_threads_divergent.min","smsp__sass_branch_targets_threads_divergent.sum","smsp__sass_inst_executed_memdesc_explicit_evict_type","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_first","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_last","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_normal","smsp__sass_inst_executed_memdesc_explicit_hitprop_evict_normal_demote","smsp__sass_inst_executed_memdesc_explicit_missprop_evict_first","smsp__sass_inst_executed_memdesc_explicit_missprop_evict_normal","smsp__sass_inst_executed_op_dshared.sum","smsp__sass_inst_executed_op_dshared.sum.pct_of_peak_sustained_elapsed","smsp__sass_inst_executed_op_dshared_atom.sum","smsp__sass_inst_executed_op_dshared_ld.sum","smsp__sass_inst_executed_op_dshared_redas.sum","smsp__sass_inst_executed_op_dshared_st.sum","smsp__sass_inst_executed_op_dshared_stas.sum","smsp__sass_inst_executed_op_dshared_tma_red.sum","smsp__sass_inst_executed_op_dshared_tma_st.sum","smsp__sass_inst_executed_op_global_ld.sum","smsp__sass_inst_executed_op_global_st.sum","smsp__sass_inst_executed_op_local_ld.sum","smsp__sass_inst_executed_op_local_st.sum","smsp__sass_inst_executed_op_shared.sum","smsp__sass_inst_executed_op_shared.sum.pct_of_peak_sustained_elapsed","smsp__sass_inst_executed_op_shared_ld.sum","smsp__sass_inst_executed_op_shared_st.sum","smsp__sass_inst_executed_op_tma_ld.sum","smsp__sass_inst_executed_op_tma_red.sum","smsp__sass_inst_executed_op_tma_st.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts.sum.pct_of_peak_sustained_elapsed","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts_cache_access.sum","smsp__sass_l1tex_data_pipe_lsu_wavefronts_mem_shared_op_ldgsts_cache_access.sum.pct_of_peak_sustained_elapsed","smsp__sass_l1tex_m_xbar2l1tex_read_sectors_mem_global_op_ldgsts_cache_bypass.sum","smsp__sass_l1tex_m_xbar2l1tex_read_sectors_mem_global_op_ldgsts_cache_bypass.sum.pct_of_peak_sustained_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dfma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_dmul_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_ffma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_fmul_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hadd_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hfma_pred_on.sum.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.avg.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.max.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.min.per_cycle_elapsed","smsp__sass_thread_inst_executed_op_hmul_pred_on.sum.per_cycle_elapsed","smsp__thread_inst_executed_per_inst_executed.ratio","smsp__thread_inst_executed_pred_on_per_inst_executed.ratio","smsp__warps_active.avg.peak_sustained","smsp__warps_active.avg.per_cycle_active","smsp__warps_active.max.peak_sustained","smsp__warps_active.max.per_cycle_active","smsp__warps_active.min.peak_sustained","smsp__warps_active.min.per_cycle_active","smsp__warps_active.sum.peak_sustained","smsp__warps_active.sum.per_cycle_active","smsp__warps_eligible.avg.per_cycle_active","smsp__warps_eligible.max.per_cycle_active","smsp__warps_eligible.min.per_cycle_active","smsp__warps_eligible.sum.per_cycle_active","thread_inst_executed","thread_inst_executed_true"
+"","","","","","","","","","","","","","thread","thread","thread","thread","","","","%","Kbyte","Tbyte","","","byte","","%","%/register","%/Kbyte","thread","thread","thread","%","thread","thread","thread","thread","thread","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","","request","%","request","%","","%","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","ms","ms","ms","ms","","","","","block","block","block","block","","","","","%","%","%","%","%","","%","inst","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","","","","","","%","%","%","%","%","%","%","%","%","%","%","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","cycle","%","","Khz","%","%","%","%","%","%","%","%","Mbyte","%","Gbyte/s","byte","%","byte/s","byte","%","byte/s","Mbyte","%","Gbyte/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector","%","sector/s","sector","%","sector/us","sector","%","sector","%","sector","%","sector","%","Gbyte","%","Tbyte/s","byte","%","byte/s","Gbyte","%","Tbyte/s","%","%","%","%","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector/s","sector","%","sector","%","sector","%","sector","%","sector/ns","sector","%","sector","%","sector","%","sector","%","byte","%","byte/s","","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","","%","%","%","%","%","%","%","%","%","%","%","%","%","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","%","%","cycle","%","%","%","%","%","%","%","%","%","%","%","%","%","","block","block","block","","","","","cluster","block","","","","","","","","","","","","%","%","block","block","block","block","block","","","","","","Mbyte","","","","","register/thread","register/thread","Kbyte","Kbyte/block","Kbyte/block","Kbyte/block","Kbyte/block","byte/block","SM","","","thread","","","","","","","","","%","sector","%","","","","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","%","%","%","%","%","%","%","%","%","%","%","sector","%","sector/ns","sector","sector","sector","sector","%","","Ghz","%","","Ghz","%","","Ghz","%","","Ghz","request","request","request","request","request","request","request","request","%","%","%","sector","%","sector/cycle","sector/cycle","sector","%","sector","%","sector","%","sector/ns","sector","%","sector/us","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/ns","%","sector/cycle","sector/cycle","sector","sector/ms","sector","sector","%","sector/cycle","sector/cycle","%","%","sector","%","sector/ns","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/s","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","%","sector/cycle","sector/cycle","sector","sector/s","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/ns","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/s","sector","sector","%","sector/cycle","sector/cycle","sector","%","sector/us","sector","sector","sector","sector","sector","sector","sector","sector","sector","sector","%","%","%","%","%","%","%","%","%","%","%","%","sector","%","sector/s","sector","%","sector/s","","","sectors","sectors","sectors","sectors","sectors","sectors","","","","","","","","","","","","","","","","","","","","","","","","block","","cycle","%","%","%","cycle","%","inst/cycle","%","%","%","warp","warp","","Mbyte","","","us","sample","","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","Gbyte","pass","pass","","","","","inst","inst","inst","inst","","","","","inst","inst","inst","inst","inst","inst","inst","","","","","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","%","cycle","%","cycle","%","cycle","%","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","inst/cycle","inst/cycle","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","inst/cycle","%","inst/cycle","%","inst/cycle","%","inst/cycle","%","%","%","%","%","%","%","%","%","%","%","%","warp","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","","%","","","","","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","%","inst","inst","byte","%","byte/s","","%","","","sector","%","sector","%","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","%","%","%","%","%","warp","%","warp","%","warp","%","warp","cycle","cycle","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","branches","cycle","cycle","cycle","cycle","cycle","Ghz","cycle","Ghz","cycle","Ghz","cycle","Ghz","inst","inst","inst","inst","inst","inst","inst","inst","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","%","inst","inst","inst","inst","%","","%","","%","","%","","%","%","%","%","warp","pass","Mbyte","byte","","cycle","","warp","warp","branches","branches","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","inst","inst","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","%","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","byte/sector","branches","branches","branches","branches","","inst","inst","inst","inst","inst","inst","inst","%","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","inst","%","inst","inst","inst","inst","inst","","%","","%","sector","%","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","inst/cycle","","","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","warp","inst","inst"
+"0","73","python3.12","127.0.0.1","cutlass3x_sm120_bstensorop_s16864gemm_block_scaled_ue4m3xe2m1_ue4m3xe2m1_f32_bf16_bf16_128x128x256_1x1x1_0_tnn_align32_o_vs16_bias_bf16_relu","1","7","(384, 1, 1)","(296, 42, 1)","0","12.1","0","1","49636","46015","95873.000000","6036","6144","7215061.44","0","0","1.024000","1.507107","293","44556288","0","625","316.000000","4225.000000","4.975000","192.000000","12288.000000","12288.000000","0.030323","0","11.216880","0","0","0","36608229","1207261415","0","0","nan","0","432","1","0","1","1","1","0","1","0","1","0","443","2418000","1","12","1","0","1","1","1","No-CC","1","1","1","0","0","NVIDIA GB10","0","0","256","4","1","1","256","1","0","0","100","1","772935902","11794","32000","16","0","4","2500","16","161","4318","0","1","0","0","1","0","1","443","1","1","1","25165824","24","24","24","1","1","134217728","1024","1024","64","24","2418000","2147483647","65535","65535","4","1","8533000","18874368","2147483647","65536","65536","255","49152","101376","102400","1024","1536","48","12","2048","32768","32768","65536","32768","2048","32768","131072","16384","16384","16384","2046","32768","32768","2048","32768","268435456","32768","131072","32768","32768","65536","32768","2048","32768","65000","2097120","131072","32768","32768","131072","16384","32768","16384","8192","16384","8192","2046","32768","32768","4","8533000","1","1","0","0","0","0","48","6","4","1","0","1","1","1","0","15","1","0","1024","12","64","1","1","512","0","1","512","32","65536","130661769216","1","1","1","32","2798220.000000","1.926207","5640.000000","0.003882","8224","0.002831","36317742.500000","2.132444","36338193.000000","2.133645","36297325.000000","2.131245","145270970.000000","8.529777","77.673393","77.716884","77.629395","77.673393","38.962587","39.005204","38.932311","38.962587","76.458743","76.606695","76.298073","76.458743","0","0","0","0","77.673393","77.716884","77.629395","77.673393","17.031040","17.031040","17.031040","17.031040","12384","12384","12384","12384","12384.000000","12384.000000","12384.000000","12384.000000","12432","12432","12432","12432","0.208009","0.210415","0.206373","0.208009","99.819427","1809792","0.207634","1180803976","36039245.812500","36113066.000000","35944597.000000","1729883799.000000","36317742.500000","2.132444","36338193.000000","2.133645","36297325.000000","2.131245","1743251640.000000","102.357322","47563205","0","46508786","0","1034202","17.344946","17.479037","17.210811","17.344946","5.842399","5.887514","5.797284","5.842399","41.880489","42.205724","41.556958","41.880489","727700427","41.743854","0","0","715093417","41.020665","4223650","0.242286","0","0","0","0","0","0","0","0","38.684431","39.132509","38.236352","38.684431","960.000000","0.000055","48","56.367668","26.632013","26.940474","26.426373","26.632013","11.606359","11.735944","11.477431","11.606359","406.683648","0.729032","23.878967","0","0","0","0","0","0","406.683648","0.729032","23.878967","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","12708864.000000","0.729032","746.217730","0","0","0","0","0","0","0","0","25.664425","46.006742","1.506921","0","0","0","25.664422","46.006736","1.506920","46.006742","46.539638","45.651477","46.006742","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","802013184.000000","46.006736","47.091263","96.000000","0.000006","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000055","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960","0.000055","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","90.000000","0","90.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","960.000000","864.000000","96.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.042018","0.042398","0.041635","0.042018","46.362263","46.006742","46.899278","46.539638","46.004253","45.651477","46.362263","46.006742","8","384","1","1","384","0","0","0","0","8","PolicySpread","0","1","0","CachePreferNone","1","296","42","1","12432","","0","0","3.000000","24.000000","1.000000","1.000000","4.000000","300","157","0","2028","2388","4.718592","0","0","0","0","168","168.000000","102.400000","89.088000","89.088000","1.024000","88.064000","0","48","1024","7","4773888","24","all","0","0","0","0","0","259","100","0","0","0","1.00","0","32545999.437500","32657096.000000","32492211.000000","520735991.000000","32783718.000000","1.924939","32783718.000000","1.924939","32783718.000000","1.924939","524539488.000000","30.799029","0","0","0","0","0","0","0","0","42.276093","42.299693","42.254622","42.276093","59982309.000000","11.435232","3.521940","0","0","0","0","76.458743","2","2.943569","76.606695","2","2.949265","76.298073","2","2.937383","76.458743","32","47.097100","203738812.000000","2176.000000","203680608.000000","0","0","200503392.000000","0","3177216.000000","91.101458","92.547312","0.149636","50928452.437500","77.673393","2.000000","1.553468","50956968.000000","77.716884","50899604.000000","77.629395","814855239.000000","77.673393","47.845301","12745592.000000","2.429863","748.374263","0","8.000000","0","0","0","19228.000000","76601.000000","0","0","0","0","742371185.000000","72417342.000000","742345007.000000","72310114.000000","802013280.000000","19.112319","47.091269","0.001659","1.000000","0.000017","8704.000000","511.066852","6648.000000","2056.000000","77.660706","2.000000","1.553214","77.704963","77.619970","814722144.000000","77.660706","47.837486","0","0","0","0","4538341.875000","4540968.000000","4534792.000000","72613470.000000","0","0","0","0","0","0","742137138.000000","72647642.000000","742437594.000000","4540477.625000","4542744.000000","4537720.000000","72647642.000000","0","0","0","0","1.000000","0","0","0","0","0","0","1.000000","0","0","0","0","0","0","0","0","0","0","0","76.449276","2.000000","1.528986","802013280.000000","76.449276","47.091269","0","0","0","0","0","0","742415110.000000","59664462.000000","742415110.000000","59664462.000000","0","1.000000","0","0","0","0","0","0","2.422861","1.000000","0.024229","12708864.000000","2.422861","746.217730","0","0","0","0","0","0","0","12708864.000000","0","12708864.000000","38.843584","38.867529","38.819355","38.843584","77.673393","77.716884","77.629395","77.673393","38.658720","38.685603","38.634477","38.658720","0","0","0","0","0","0","0","0","960","716319840","671763552","960","960","0","0","1","1","1","1","0","0","0","0","0","0","0","0","1","0","0","0","0","0","0","0","0","0","32.000000","15209492.98","14049312.000000","","","","34583650.437500","3.723238","1969.697337","84.262067","2.030255","0.099917","27009.964736","1296478.307310","0","18.022400","1","1","6.000000","0","2","25.185856","25.185856","25.185856","957.062541","25.185856","25.185856","25.185856","957.062541","38.000000","0","1","1","1","1","1000.000000","0","0","0","1180803976","1180803976","1180803976","1180803976","0","no data","no data","0","0","89236896.000000","450.000000","29478400372","29478400372","29478400372","29478400372","36039245.812500","36113066.000000","35944597.000000","1729883799.000000","36317742.500000","2.132444","36338193.000000","2.133645","36297325.000000","2.131245","1743251640.000000","102.357322","99.463631","816788.000000","0.046854","812407.000000","99.970991","36381373.000000","4.173967","17.313356","0.697886","0.692534","17.505091","0.705615","0.700204","17.184602","0.692696","0.687384","17.313356","33.498521","33.241643","4.263262","4.230570","4.283103","4.250259","4.246226","4.213665","4.263262","4.230570","3.641974","3.614047","3.684159","3.655907","3.613817","3.586105","3.641974","3.614047","0.146173","0.145052","0.146973","0.145846","0.145061","0.143948","0.146173","0.145052","0.520434","0.527147","0.511312","0.520434","0.531922","0.527843","0.536057","0.531947","0.527794","0.523747","0.531922","0.527843","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","26.837815","26.632013","27.045044","26.837654","26.630585","26.426373","26.837815","26.632013","82.421791","81.789753","0","0","0","0","0","0","0","0","0","0","0.517478","0.513509","0.523471","0.519457","0.513482","0.509544","0.517478","0.513509","3.397567","3.371514","3.423808","3.397554","3.371328","3.345476","3.397567","3.371514","0.000719","0.000713","0.000727","0.000721","0.000713","0.000708","0.000719","0.000713","0","0","0","0","0","0","0","0","17.420154","0.702191","17.611351","0.709898","17.289200","0.696912","17.420154","33.705158","81.789753","82.421334","80.842382","81.789753","0","0","0","0","17.420154","17.611351","17.289200","17.420154","12.000000","25.000000","26.632013","26.837654","26.426373","26.632013","0.046850","0.047387","0.046492","0.046850","9.697062","9.809381","9.584743","9.697062","10.401641","10.478072","10.326578","10.401641","3.720293","3.751585","3.677941","3.720293","3.720318","3.751610","3.677965","3.720318","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","49152","104813897410845.14","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","49152","104813897410845.14","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","49152","104813897410845.14","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","512","1091811431362.97","0","0","0","0","512","1091811431362.97","0","0","0","0","512","1091811431362.97","0","0","0","0","24576","52406948705422.57","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","49152","104813897410845.14","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","121668370432","81.789753","4096","8734491450903.76","3350.11","7143919010935.33","122607894528","82.421334","4096","8734491450903.76","3375.98","7199084408703.17","120728846336","81.158172","4096","8734491450903.76","3324.24","7088753613167.49","5840081780736","81.789753","196608","419255589643380.56","160805.20","342908112524895.69","0","0","8192","17468982901807.52","0","0","0","0","8192","17468982901807.52","0","0","0","0","8192","17468982901807.52","0","0","0","0","393216","838511179286761.12","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","3.51","7478160488.79","0","0","0","0","3.51","7478160488.79","0","0","0","0","3.51","7478160488.79","0","0","0","0","168.33","358951703461.80","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","2048","4367245725451.88","0","0","0","0","98304","209627794821690.28","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","4096","8734491450903.76","0","0","0","0","196608","419255589643380.56","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","1024","2183622862725.94","0","0","0","0","49152","104813897410845.14","0","0","3.641990","3.614062","3.670124","3.641980","3.601820","3.574200","3.641990","3.614062","6.216056","6.264179","6.143976","6.216056","0.676900","0.671709","0.681537","0.676310","0.672651","0.667493","0.676900","0.671709","1.999371","2.014814","1.976159","1.999371","0","0","0","0","0","0","0","0","82.421791","81.789753","83.058253","82.421334","81.467099","80.842382","82.421791","81.789753","82.421791","81.789753","0","0","0.517478","0.513509","0.523471","0.519457","0.513482","0.509544","0.517478","0.513509","0","0","0","0","0","0","0","0","0","0","0","0","0","2.000000","2.000000","2.000000","96.000000","128.000000","128.000000","128.000000","6144.000000","64.000000","64.000000","64.000000","3072.000000","81.789753","82.421334","80.842382","81.789753","20.956326","10.059036","21.002397","10.081151","20.916088","10.039722","20.956326","482.833740","13.603819","13.687736","0.172007","0.158464","0.043709","0.000001","0","0.601708","4.064508","0","0.049404","0.014477","0.071353","0.306477","1.000138","0.047087","3.751652","0","3.981793","12480","37642027.255208","37727010.000000","37561660.000000","7227269233.000000","36317742.500000","2.132444","36338193.000000","2.133645","36297325.000000","2.131245","6973006560.000000","409.429287","6287819.869792","6850658.000000","6041254.000000","1207261415.000000","190667.859375","252741.000000","152368.000000","36608229.000000","0","0","0","0","0","0","0","0","133668864.000000","1.916947","0","0","1591296.000000","0.022821","0","0","0","0","0","0","0","0","0","0","0","0","2784768.000000","0.039936","99456.000000","0.001426","6326606.692708","6912027.000000","6063398.000000","1214708485.000000","16.807295","0.17","18.362526","0.18","16.108054","0.16","16.807295","32.27","83.192705","81.863240","83.678442","83.192705","3.000000","2","33.554432","0","5","1024.000000","1619549","24610","21842","18275","15459","5250","3226","1","0","0","0","69715","61673","503007","433438","0","0","5470","4938","1697","1188","8151","6640","37460","0","122716","0","6196","5230","326651","280297","0","0","490350","479775","99.927449","32.000000","4.000000","32.000000","0","32.000000","0","32.000000","0","65.010417","336.000000","0","12482.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","1000.000000","0","0","0","224497506.000000","3.219522","89236896.000000","450.000000","2784768.000000","0","99456.000000","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0.029211","0.037782","0.022331","5.608440","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","0","31.21","25.05","12.000000","2.286434","12.000000","2.749732","12.000000","1.825086","2304.000000","438.995350","0.219602","0.242512","0.209829","42.163582","36861708726","29478400372"
diff --git a/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep b/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep
new file mode 100644
index 0000000..1ed55b6
Binary files /dev/null and b/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep differ
diff --git a/benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json b/benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json
new file mode 100644
index 0000000..0264421
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json
@@ -0,0 +1,1741 @@
+{
+  "mode": "profile",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 12,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 32,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "split_k": 1,
+    "reduction_scheme": 0,
+    "custom_option": 0,
+    "cta_swizzle": 0
+  },
+  "profile": {
+    "target": "candidate",
+    "selected_config": {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0
+    },
+    "supplied_workspace_bytes": 0,
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json b/benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json
new file mode 100644
index 0000000..e5499d6
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json
@@ -0,0 +1,1769 @@
+{
+  "mode": "trajectory",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 12,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 32,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "split_k": 1,
+    "reduction_scheme": 0,
+    "custom_option": 0,
+    "cta_swizzle": 0
+  },
+  "trajectory": {
+    "steps": 12,
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "seed": 440420,
+    "baseline_seconds": 278.20053176092915,
+    "candidate_seconds": 255.37142011406831,
+    "improvement_percent": 8.205991376924814,
+    "video_parity": {
+      "bf16_exact": true,
+      "different_elements": 0,
+      "max_abs": 0.0,
+      "mean_abs": 0.0,
+      "actual_sha256": "c62d23a42972eab907ba42f93c50247ff17a9c454b4a53fe93d2e34f9fefe578",
+      "expected_sha256": "c62d23a42972eab907ba42f93c50247ff17a9c454b4a53fe93d2e34f9fefe578"
+    },
+    "audio_parity": {
+      "bf16_exact": true,
+      "different_elements": 0,
+      "max_abs": 0.0,
+      "mean_abs": 0.0,
+      "actual_sha256": "852005383770480a6503504e1ffec86dd1fb63a69c6400f92da18e39e0986de2",
+      "expected_sha256": "852005383770480a6503504e1ffec86dd1fb63a69c6400f92da18e39e0986de2"
+    },
+    "bf16_exact": true,
+    "selected_config": {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0
+    },
+    "supplied_workspace_bytes": 0,
+    "all_50_fc2_calls_replaced": true,
+    "accepted_swiglu_producer_preserved": true,
+    "accepted_gate_and_residual_path_preserved": true
+  },
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json b/benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json
new file mode 100644
index 0000000..364fd0a
--- /dev/null
+++ b/benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json
@@ -0,0 +1,1769 @@
+{
+  "mode": "trajectory",
+  "environment": {
+    "platform": "Linux-6.17.0-1026-nvidia-aarch64-with-glibc2.39",
+    "python": "3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0]",
+    "torch": "2.9.1+cu130",
+    "torch_cuda": "13.0",
+    "device": "NVIDIA GB10",
+    "device_capability": [
+      12,
+      1
+    ],
+    "driver": null,
+    "git_commit": null,
+    "checkpoint_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors",
+    "checkpoint_sha256": null,
+    "checkpoint_hash_note": "not calculated",
+    "environment_switches": {
+      "CUDA_DEVICE_MAX_CONNECTIONS": "1",
+      "CUDA_DEVICE_MAX_COPY_CONNECTIONS": "4",
+      "CUDA_HOME": "/usr/local/cuda",
+      "CUDA_INC_PATH": "/usr/local/cuda/include",
+      "CUDA_MANAGED_FORCE_DEVICE_ALLOC": "1",
+      "CUDA_MODULE_LOADING": "EAGER",
+      "CUDA_VERSION": "13.0.2",
+      "H3_FUSED_ELEMENTWISE": "1",
+      "H3_MODEL_PATH": "/models/minimax_h3_ref2va_pruned_nvfp4.safetensors",
+      "H3_NVFP4_MODULATE_FUSION": "1",
+      "H3_NVFP4_SCALE_BACKEND": "vortex",
+      "H3_NVFP4_SCALE_VERSION": "1",
+      "H3_NVFP4_SWIGLU_FUSION": "1",
+      "H3_SAGE_QKV_LAYOUT": "strided_nhd",
+      "TORCH_COMPILE_DISABLE": "0",
+      "TORCH_CUDA_ARCH_LIST": "12.1a",
+      "TORCH_EXTENSIONS_DIR": "/opt/h3-blackwell-runtime/.torch_extensions"
+    },
+    "extension": {
+      "cuda_version": 13000,
+      "cublas_version": 130100,
+      "stream_k_public_control": false,
+      "stream_k_note": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    },
+    "comfy_kitchen": "0.2.31 package without __version__"
+  },
+  "workload": {
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "steps": 2,
+    "sampler_step": 1,
+    "seed": 440420,
+    "text_tokens": 100,
+    "tokens": 37810,
+    "hidden_shape": [
+      37810,
+      5376
+    ],
+    "segments": [
+      [
+        0,
+        100,
+        1
+      ],
+      [
+        100,
+        514,
+        2
+      ],
+      [
+        514,
+        37810,
+        0
+      ]
+    ]
+  },
+  "retained_blocks": [
+    0,
+    24,
+    49
+  ],
+  "immutable_cloned_block_inputs": {
+    "0": [
+      37810,
+      5376
+    ],
+    "24": [
+      37810,
+      5376
+    ],
+    "49": [
+      37810,
+      5376
+    ]
+  },
+  "fc2_boundary": {
+    "block": 24,
+    "gate_up_shape": [
+      37810,
+      28672
+    ],
+    "activation_qdata_shape": [
+      37824,
+      7168
+    ],
+    "weight_qdata_shape": [
+      5376,
+      7168
+    ],
+    "logical_mnk": [
+      37810,
+      5376,
+      14336
+    ],
+    "descriptor_mnk_after_padding": [
+      37824,
+      5376,
+      14336
+    ],
+    "producer": "vortex_native_quantize_swiglu_nvfp4",
+    "no_bias": true
+  },
+  "baseline_kernel_metadata": {
+    "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+    "descriptors": {
+      "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+      "block_scale_mode": "VEC16_UE4M3",
+      "compute_and_scale": "FP32",
+      "scalar_pointer_mode": "device",
+      "bias": null,
+      "beta": 0.0,
+      "comfy_kitchen_version": "0.2.31"
+    },
+    "profiler_cuda_events_available": false,
+    "profiler_note": "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+    "top_cuda_events": [],
+    "output_sha256": "ad26c744afbb0249bff9afea503984e51217f85a59434e44fde347ae7d9b852f"
+  },
+  "heuristics": [
+    {
+      "max_workspace_bytes": 0,
+      "requested_count": 32,
+      "returned_count": 5,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 4194304,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 8388608,
+      "requested_count": 32,
+      "returned_count": 7,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 16777216,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 33554432,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    },
+    {
+      "max_workspace_bytes": 67108864,
+      "requested_count": 32,
+      "returned_count": 6,
+      "algorithms": [
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": -2,
+          "reduction_scheme": 2,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        },
+        {
+          "algorithm_id": 70,
+          "tile_id": 20,
+          "stages_id": 37,
+          "split_k": 1,
+          "reduction_scheme": 0,
+          "custom_option": 0,
+          "cta_swizzle": 0,
+          "inner_shape": null,
+          "cluster_shape": null,
+          "required_workspace_bytes": 0,
+          "waves": 1.0,
+          "state": 0,
+          "api_status": 0,
+          "valid": true,
+          "capabilities": {
+            "split_k_support": 1,
+            "reduction_scheme_mask": 6,
+            "cta_swizzle_support": 0,
+            "custom_option_max": 0,
+            "strided_batch_support": 1,
+            "out_of_place_result_support": 1,
+            "tile_ids": [
+              20
+            ],
+            "stages_ids": [
+              37
+            ],
+            "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+          }
+        }
+      ]
+    }
+  ],
+  "explicit_split_k_checks": [
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 1.0,
+      "state": 0,
+      "api_status": 0,
+      "valid": true,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 1,
+        "reduction_scheme": 0
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 2,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 2,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 4,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 4,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 8,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 8,
+        "reduction_scheme": 4
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 2,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 2
+      }
+    },
+    {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 16,
+      "reduction_scheme": 4,
+      "custom_option": 0,
+      "cta_swizzle": 0,
+      "inner_shape": null,
+      "cluster_shape": null,
+      "required_workspace_bytes": 0,
+      "waves": 0.0,
+      "state": 15,
+      "api_status": 15,
+      "valid": false,
+      "capabilities": {
+        "split_k_support": 1,
+        "reduction_scheme_mask": 6,
+        "cta_swizzle_support": 0,
+        "custom_option_max": 0,
+        "strided_batch_support": 1,
+        "out_of_place_result_support": 1,
+        "tile_ids": [
+          20
+        ],
+        "stages_ids": [
+          37
+        ],
+        "inner_cluster_shape_capability_note": "This CUDA 13 cublasLt.h exposes config IDs but no public capability attributes that enumerate inner/cluster shape IDs."
+      },
+      "requested_config": {
+        "algorithm_id": 70,
+        "tile_id": 20,
+        "stages_id": 37,
+        "custom_option": 0,
+        "cta_swizzle": 0,
+        "inner_shape": null,
+        "cluster_shape": null,
+        "split_k": 16,
+        "reduction_scheme": 4
+      }
+    }
+  ],
+  "selected": {
+    "algorithm_id": 70,
+    "tile_id": 20,
+    "stages_id": 37,
+    "split_k": 1,
+    "reduction_scheme": 0,
+    "custom_option": 0,
+    "cta_swizzle": 0
+  },
+  "trajectory": {
+    "steps": 2,
+    "resolution": [
+      1344,
+      768
+    ],
+    "frames": 124,
+    "seed": 440420,
+    "baseline_seconds": 45.9624265920138,
+    "candidate_seconds": 42.51584641600493,
+    "improvement_percent": 7.49869062963644,
+    "video_parity": {
+      "bf16_exact": true,
+      "different_elements": 0,
+      "max_abs": 0.0,
+      "mean_abs": 0.0,
+      "actual_sha256": "7d5e50d8f0ee9a639feede8a0e6aea1bd160087634ecc7ae80d0ef5b96fd8e78",
+      "expected_sha256": "7d5e50d8f0ee9a639feede8a0e6aea1bd160087634ecc7ae80d0ef5b96fd8e78"
+    },
+    "audio_parity": {
+      "bf16_exact": true,
+      "different_elements": 0,
+      "max_abs": 0.0,
+      "mean_abs": 0.0,
+      "actual_sha256": "df518aa8c645ec45e7cc46be58a941790f9766b420adf2cdcbaf14ef8b0e5605",
+      "expected_sha256": "df518aa8c645ec45e7cc46be58a941790f9766b420adf2cdcbaf14ef8b0e5605"
+    },
+    "bf16_exact": true,
+    "selected_config": {
+      "algorithm_id": 70,
+      "tile_id": 20,
+      "stages_id": 37,
+      "split_k": 1,
+      "reduction_scheme": 0,
+      "custom_option": 0,
+      "cta_swizzle": 0
+    },
+    "supplied_workspace_bytes": 0,
+    "all_50_fc2_calls_replaced": true,
+    "accepted_swiglu_producer_preserved": true,
+    "accepted_gate_and_residual_path_preserved": true
+  },
+  "errors_and_unsupported": [
+    {
+      "feature": "Stream-K",
+      "supported_public_control": false,
+      "reason": "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute that directly selects Stream-K. Negative SPLITK_NUM values returned by heuristics are preserved as undocumented library sentinels, not claimed as public Stream-K control."
+    }
+  ]
+}
diff --git a/research/artifact_manifest.json b/research/artifact_manifest.json
index b916e25..2dfd7f2 100644
--- a/research/artifact_manifest.json
+++ b/research/artifact_manifest.json
@@ -2,39 +2,199 @@
   "metadata": {
     "algorithm": "SHA-256",
     "generated_date": "2026-08-25",
-    "scope_note": "Local selected scope is the established 167-file set. Spark records are the complete reproducible current top-level benchmark output set; the audit-retained 279-file/665950155-byte aggregate cannot be reconstructed because its path list was not retained.",
+    "scope_note": "Local selected scope is the established 167-file set plus 10 retained FC2 scheduling artifacts. Spark records are the complete reproducible current top-level benchmark output set; the audit-retained 279-file/665950155-byte aggregate cannot be reconstructed because its path list was not retained.",
     "summary": {
       "local": {
-        "record_count": 167,
-        "size_bytes": 210388688,
-        "expected_record_count": 167,
-        "expected_size_bytes": 210388688,
+        "record_count": 177,
+        "size_bytes": 230157548,
+        "expected_record_count": 177,
+        "expected_size_bytes": 230157548,
         "reconciled": true
       },
       "spark": {
-        "record_count": 280,
-        "size_bytes": 547708583,
+        "record_count": 290,
+        "size_bytes": 567477443,
         "expected_record_count": 279,
         "expected_size_bytes": 665950155,
         "reconciled": false,
-        "record_count_delta": 1,
-        "size_bytes_delta": -118241572
+        "record_count_delta": 11,
+        "size_bytes_delta": -98472712
       },
       "total": {
-        "record_count": 447,
-        "size_bytes": 758097271
+        "record_count": 467,
+        "size_bytes": 797634991
       }
     },
     "json_reconciliation": {
-      "identical": 106,
+      "identical": 112,
       "mismatches": 2,
       "local_only": 40,
       "spark_only": 107,
-      "local_total": 148,
-      "spark_total": 215
+      "local_total": 154,
+      "spark_total": 221
     }
   },
   "artifacts": [
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv",
+      "size_bytes": 164324,
+      "sha256": "e5bfd39f1512737c0852975de319b502e2ce20677e2a51ad0ce50656923634a5",
+      "artifact_class": "nsight_csv_export",
+      "location": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep",
+      "size_bytes": 13364977,
+      "sha256": "975eea7110a19b472ef8cd29ec1c72f629b0f97e224080acc50a43b371bc290d",
+      "artifact_class": "nsight_compute_report",
+      "location": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json",
+      "size_bytes": 51924,
+      "sha256": "082e195a34a0a4738cd515dcc7db69e59d8bef22afcb49e73a91d21cb0ccaff0",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json",
+      "size_bytes": 61451,
+      "sha256": "d1051463765052c83494f417b40b9fe9eb389eb4ebf213ff7ff491991d569d8c",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json",
+      "size_bytes": 64494,
+      "sha256": "c762e9394f6ab6dd1edfd13097a7fa242caa4540fb3d71c3b1035b3edbcd3261",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv",
+      "size_bytes": 148497,
+      "sha256": "2a9da47628571fd430a8c0ba4fce3547d8e12dec3b443b101a3a8baa4bdf8e8f",
+      "artifact_class": "nsight_csv_export",
+      "location": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep",
+      "size_bytes": 5755084,
+      "sha256": "b368a14dcd96f859dbad9e7a8fa10bc33bf6373dbea8972b6685a2d83b68dfae",
+      "artifact_class": "nsight_compute_report",
+      "location": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json",
+      "size_bytes": 52089,
+      "sha256": "7b2282b6381e90d07d7175fe3a8f1aa03c7d85a3c6d25cab2350abf6559800d8",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json",
+      "size_bytes": 53013,
+      "sha256": "075c3aecbe6f77cc28f07a8f656de044e437c94a9fc71068c0ac9e8f3082972c",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json"
+    },
+    {
+      "scope": "local",
+      "path": "benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json",
+      "size_bytes": 53007,
+      "sha256": "a9cc0b229dcc7e37705d94705b1e607092d9aee7e3873787eff539074f46228f",
+      "artifact_class": "benchmark_json",
+      "location": "benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv",
+      "size_bytes": 164324,
+      "sha256": "e5bfd39f1512737c0852975de319b502e2ce20677e2a51ad0ce50656923634a5",
+      "artifact_class": "nsight_csv_export",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep",
+      "size_bytes": 13364977,
+      "sha256": "975eea7110a19b472ef8cd29ec1c72f629b0f97e224080acc50a43b371bc290d",
+      "artifact_class": "nsight_compute_report",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json",
+      "size_bytes": 51924,
+      "sha256": "082e195a34a0a4738cd515dcc7db69e59d8bef22afcb49e73a91d21cb0ccaff0",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json",
+      "size_bytes": 61451,
+      "sha256": "d1051463765052c83494f417b40b9fe9eb389eb4ebf213ff7ff491991d569d8c",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json",
+      "size_bytes": 64494,
+      "sha256": "c762e9394f6ab6dd1edfd13097a7fa242caa4540fb3d71c3b1035b3edbcd3261",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv",
+      "size_bytes": 148497,
+      "sha256": "2a9da47628571fd430a8c0ba4fce3547d8e12dec3b443b101a3a8baa4bdf8e8f",
+      "artifact_class": "nsight_csv_export",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep",
+      "size_bytes": 5755084,
+      "sha256": "b368a14dcd96f859dbad9e7a8fa10bc33bf6373dbea8972b6685a2d83b68dfae",
+      "artifact_class": "nsight_compute_report",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json",
+      "size_bytes": 52089,
+      "sha256": "7b2282b6381e90d07d7175fe3a8f1aa03c7d85a3c6d25cab2350abf6559800d8",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json",
+      "size_bytes": 53013,
+      "sha256": "075c3aecbe6f77cc28f07a8f656de044e437c94a9fc71068c0ac9e8f3082972c",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json"
+    },
+    {
+      "scope": "spark",
+      "path": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json",
+      "size_bytes": 53007,
+      "sha256": "a9cc0b229dcc7e37705d94705b1e607092d9aee7e3873787eff539074f46228f",
+      "artifact_class": "benchmark_json",
+      "location": "/home/daniel/StoryStudioAssets/H3-output/h3-blackwell-runtime/benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json"
+    },
     {
       "scope": "local",
       "path": "benchmarks/audio-dialogue-format-sweep-sage2-seeds440420-440429.json",
diff --git a/research/experiment_registry.json b/research/experiment_registry.json
index ffd85da..ff061b6 100644
--- a/research/experiment_registry.json
+++ b/research/experiment_registry.json
@@ -500,6 +500,28 @@
       "reproducer_commands": [], "timestamp": null, "evidence_missing": ["No accepted alternative FC2 reduction implementation"],
       "production_behavior": "FC2 retains the Comfy/CUBLAS fallback.", "source_recovery": "P0 artifacts and fallback policy are retained in the roadmap."
     },
+    {
+      "id": "fc2-cublaslt-splitk1-schedule",
+      "name": "FC2 cuBLASLt public split-K-1 schedule",
+      "family": "nvfp4-library-scheduling",
+      "status": "research_retained",
+      "hypothesis": "A documented cuBLASLt schedule can preserve the exact FC2 reduction result while avoiding the production heuristic's traffic and synchronization regression.",
+      "implementation_strategy": "Reproduce the exact Comfy Kitchen descriptors in an isolated extension, enumerate checked cuBLASLt algorithms, and compare one selected public split-K-1 schedule against the accepted FC2 path.",
+      "source_locations": ["research/fc2_nvfp4_scheduling/README.md", "research/fc2_nvfp4_scheduling/RESULTS.md", "research/fc2_nvfp4_scheduling/fc2_nvfp4_lt.cpp", "tools/benchmark_fc2_nvfp4_algorithms.py"],
+      "active_source_location": "research/fc2_nvfp4_scheduling/fc2_nvfp4_lt.cpp",
+      "commit_hash": null,
+      "benchmark_artifacts": [{"path": "benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json", "exists": true}],
+      "profiler_artifacts": [{"path": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv", "exists": true}, {"path": "benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv", "exists": true}],
+      "environment": {"gpu": "NVIDIA GB10", "cuda": "13", "driver": null, "pytorch": null, "triton": null, "container_image": "sha256:1d340e14cb6fc45ccfdbe63dde8db2a2b3aea94b493702c8a08e1f8d5b4f7b83", "commit_context": "isolated extension; production source and dispatch unchanged"},
+      "metrics": {"fc2_p50_ms_baseline": 53.618, "fc2_p50_ms_candidate": 15.636, "block_improvement_pct": [8.16, 8.63, 8.88], "two_step_improvement_pct": 7.50, "canonical_12_step_improvement_pct": 8.21, "trajectory_measurement_note": "single baseline-then-candidate paired runs", "l2_hit_rate_pct_baseline": 53.32, "l2_hit_rate_pct_candidate": 91.10, "l2_read_miss_reduction_bytes": 9853094784},
+      "correctness_evidence": ["Selected FC2 output is byte-exact.", "Blocks 0, 24, and 49 are byte-exact against paired baseline and retained traversal.", "Two-step and canonical 12-step video and audio latents are byte-exact."],
+      "decision_rationale": "The library schedule passed all exactness and performance gates, explaining the prior traffic amplification without requiring a custom kernel. It is retained pending explicit production integration and deployment validation.",
+      "reproducer_commands": ["python tools/benchmark_fc2_nvfp4_algorithms.py --mode sweep --rounds 20 --workspace-bytes 67108864", "python tools/benchmark_fc2_nvfp4_algorithms.py --mode block-gate --candidate research/fc2_nvfp4_scheduling/candidate_splitk1.json --rounds 20 --workspace-bytes 0"],
+      "timestamp": "2026-08-25",
+      "evidence_missing": ["Production Nvfp4Linear integration", "Production deployment smoke", "Portable validation outside GB10/SM121"],
+      "production_behavior": "Not dispatched. Production still uses Comfy Kitchen's heuristic-selected FC2 path.",
+      "source_recovery": "The direct cuBLASLt extension, benchmark harness, selected candidate, raw NCU reports, and full gate artifacts are retained in this checkout."
+    },
     {
       "id": "layout-direct-temporal-output",
       "name": "Direct-to-temporal output",
diff --git a/research/fc2_nvfp4_scheduling/README.md b/research/fc2_nvfp4_scheduling/README.md
new file mode 100644
index 0000000..800a43c
--- /dev/null
+++ b/research/fc2_nvfp4_scheduling/README.md
@@ -0,0 +1,206 @@
+# FC2 NVFP4 Library Scheduling
+
+This directory contains the first isolated library-scheduling study for H3 FC2.
+It does not register a PyTorch operator, alter `Nvfp4Linear`, or participate in
+production dispatch. The extension is loaded only by
+`tools/benchmark_fc2_nvfp4_algorithms.py`.
+
+The completed measurements and decision are in `RESULTS.md`. Algorithm 70 with
+public split-K 1 is byte-exact and passes the block, two-step, and canonical
+12-step gates, but remains research-only until production integration and
+deployment validation are performed.
+
+## Exact Operation
+
+The extension reproduces Comfy Kitchen 0.2.31's
+`cublas_gemm_nvfp4.cu` descriptors for row-major packed activation `[M,K]`
+times packed weight `[N,K]` to BF16 `[M,N]`:
+
+- cuBLASLt sees column-major `weight.T @ activation`, so Lt `m=N`, `n=M`, and
+  `k=K`; the output storage is the row-major `[M,N]` view.
+- A and B are `CUDA_R_4F_E2M1`, with
+  `CUBLASLT_MATMUL_MATRIX_SCALE_VEC16_UE4M3` block scales.
+- Compute, scalar scale type, alpha, and beta are FP32. Alpha and zero beta are
+  device pointers.
+- Output is BF16, epilogue is default, and there is no bias.
+- The activation producer is the deployed, accepted
+  `vortex_native_quantize_swiglu_nvfp4`. Alpha is its FP32 tensor scale times
+  FC2's FP32 weight tensor scale.
+
+The canonical 37,810-row input pads to 37,824 rows in the accepted producer.
+The report distinguishes logical `[M,N,K]` from descriptor/padded dimensions;
+comparison slices back to the logical output exactly as Comfy Kitchen does.
+
+## Current Baseline
+
+The accepted production path fuses exact BF16 SwiGLU into NVFP4 production and
+retains Comfy Kitchen's cuBLASLt FC2 GEMM. Existing validated metadata reports:
+
+- Producer tensor scale, QDATA, and SFA are byte-identical at blocks 0, 24, and
+  49.
+- Canonical block FC2 input is logically `37,810 x 14,336` BF16 before packing.
+- Fully fused block-24 profiling attributes 26.87% of kernel time to all four
+  NVFP4 GEMMs; this is not claimed as an FC2-only percentage.
+- Spark's accepted image is
+  `sha256:1d340e14cb6fc45ccfdbe63dde8db2a2b3aea94b493702c8a08e1f8d5b4f7b83`
+  and enables `H3_NVFP4_SCALE_BACKEND=vortex`,
+  `H3_FUSED_ELEMENTWISE=1`, `H3_NVFP4_MODULATE_FUSION=1`, and
+  `H3_NVFP4_SWIGLU_FUSION=1`.
+
+Every run also records a fresh profiler-derived baseline kernel list. Kernel
+names and times in that list are run metadata, not hard-coded claims.
+
+## Commands
+
+Run inside image `1d340e14cb6f...` from the Spark checkout, without starting the
+hot service. Preserve its production environment switches, for example:
+
+```bash
+export H3_NVFP4_SCALE_BACKEND=vortex H3_NVFP4_SCALE_VERSION=1
+export H3_FUSED_ELEMENTWISE=1 H3_NVFP4_MODULATE_FUSION=1
+export H3_NVFP4_SWIGLU_FUSION=1 H3_SAGE_QKV_LAYOUT=strided_nhd
+```
+
+Compile only:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode compile --verbose-build \
+  --build-directory /tmp/fc2-nvfp4-build
+```
+
+Default characterization enumerates workspace budgets 0, 4, 8, 16, 32, and
+64 MiB, checks a bounded set of public split-K configurations, and times one
+valid probe with alternating AB/BA order:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode characterize \
+  --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-characterize.json
+```
+
+The same workload can come from an existing capture:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode characterize \
+  --capture /artifacts/capture --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-capture.json
+```
+
+A bounded timing sweep defaults to eight candidates, not a full combinatorial
+gate:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode sweep \
+  --candidate-limit 8 --rounds 6 --workspace-bytes 33554432 \
+  --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-sweep.json
+```
+
+Re-run one selected candidate. `--candidate` accepts an inline JSON object, a
+JSON file containing one object, or `enumerate:<budget-index>:<result-index>`:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode selected \
+  --candidate '{"algorithm_id":23,"tile_id":42,"stages_id":35,"split_k":1,"reduction_scheme":0}' \
+  --workspace-bytes 33554432 --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-selected.json
+```
+
+NCU capture uses `cudaProfilerStart/Stop`; only the selected FC2 cuBLASLt call
+is inside the range:
+
+```bash
+ncu --target-processes all --profile-from-start off --set full \
+  --export /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-selected \
+  python tools/benchmark_fc2_nvfp4_algorithms.py --mode profile \
+  --candidate /tmp/fc2-candidate.json --workspace-bytes 33554432 \
+  --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-ncu.json
+```
+
+Complete block gate for blocks 0, 24, and 49:
+
+```bash
+python tools/benchmark_fc2_nvfp4_algorithms.py --mode block-gate \
+  --candidate /tmp/fc2-candidate.json --rounds 6 \
+  --workspace-bytes 33554432 --build-directory /tmp/fc2-nvfp4-build \
+  --output /output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-block-gate.json
+```
+
+Block-gate monkeypatches only `block.mlp.fc2.forward_swiglu`. It still invokes
+the accepted SwiGLU producer on the actual FC1 result. Norm, FC1, modulation,
+gate, residual, attention, and all other code remain the accepted path.
+
+## Requested NCU Traffic Counters
+
+Use `ncu --query-metrics` on the installed NCU version before replacing `--set
+full` with an explicit list. Request the available equivalents of:
+
+```text
+gpu__time_duration.sum
+dram__bytes_read.sum
+dram__bytes_write.sum
+lts__t_bytes.sum
+lts__t_sector_hit_rate.pct
+sm__throughput.avg.pct_of_peak_sustained_elapsed
+smsp__inst_executed.sum
+smsp__pipe_tensor_cycles_active.avg.pct_of_peak_sustained_active
+launch__registers_per_thread
+launch__shared_mem_per_block_allocated
+```
+
+Metric spelling and availability vary with NCU and GB10. Missing counters must
+be reported as unavailable rather than silently substituted. Also retain
+kernel duration, grid/cluster dimensions, achieved occupancy, waves, required
+workspace, and the extension's tile/stage/inner/cluster configuration.
+
+## Validation Ladder
+
+1. Compile against the image's CUDA 13 headers and print `build_info()`.
+2. Characterize heuristics and verify every executed candidate passes
+   `cublasLtMatmulAlgoCheck` within caller-supplied workspace.
+3. Compare direct-probe candidate FC2 against a cloned
+   `fc2.forward_swiglu(gate_up)` output byte-for-byte.
+4. Alternate baseline/candidate AB and BA each round; record p50, p95, dense
+   TFLOP/s, producer-plus-FC2 time, workspace, and output SHA-256.
+5. NCU one FC2 call and inspect traffic/resource counters.
+6. Gate complete blocks 0, 24, and 49 with only FC2 GEMM replaced. Require
+   BF16 exactness against both the paired baseline and retained traversal.
+7. Only after those gates should a separate, explicitly authorized trajectory
+   experiment be considered.
+
+That trajectory authorization was granted for the retained candidate. Both the
+two-step and canonical 12-step gates passed with byte-exact video and audio
+latents; see `RESULTS.md` and the linked benchmark artifacts.
+
+## Prohibited Experiments
+
+- Do not alter production source, configuration, dispatch, or existing
+  research files.
+- Do not expose this extension through `Nvfp4Linear` or any production operator.
+- Do not start or perturb the hot service.
+- Do not add bias, accumulation, host scalars, another output dtype, a different
+  producer, or approximate validation to this study.
+- Do not claim a checkpoint hash unless `--checkpoint-sha256` actually computes
+  it.
+- Do not label undocumented behavior Stream-K. CUDA 13 exposes no documented
+  public `cublasLtMatmulAlgoConfig` attribute that directly selects Stream-K;
+  the report states that limitation explicitly.
+- Do not promote a direct-GEMM result without the complete block gate and later
+  separately authorized trajectory validation.
+
+## Limitations
+
+- Heuristics are library, driver, GPU, shape, and workspace specific.
+- Explicit split-K is attempted only through documented `AlgoInit`,
+  `ConfigSet`, and `AlgoCheck`. Unsupported factors/reduction schemes are
+  recorded; they are not emulated.
+- Negative `SPLITK_NUM` values returned by the heuristic are retained as raw
+  signed library sentinels. They are not interpreted or labeled as Stream-K.
+- Characterization deliberately bounds explicit checks and timing candidates.
+  Increase limits consciously because each full canonical FC2 call is costly.
+- Caller-owned output and workspace are reused during timing. The accepted
+  producer still owns its quantized activation allocations.
+- A capture supplies packed-denoiser inputs, not pre-captured FC2 operands; the
+  harness traverses the loaded H3 model once to construct exact current
+  boundaries.
diff --git a/research/fc2_nvfp4_scheduling/RESULTS.md b/research/fc2_nvfp4_scheduling/RESULTS.md
new file mode 100644
index 0000000..761bd1b
--- /dev/null
+++ b/research/fc2_nvfp4_scheduling/RESULTS.md
@@ -0,0 +1,158 @@
+# FC2 NVFP4 Scheduling Results
+
+Status: research retained; validated as a canonical-shape production-integration
+candidate, but not wired into production dispatch.
+
+## Decision
+
+The canonical GB10 FC2 slowdown is caused by the cuBLASLt heuristic selecting a
+`_stream_k` kernel with poor traversal locality and heavy synchronization
+polling. A documented public split-K-1 configuration is byte-exact and removes
+the regression. Because this library candidate passed the complete validation
+ladder, the custom persistent-kernel branch is closed.
+
+The retained candidate is:
+
+| Field | Value |
+| --- | ---: |
+| Algorithm ID | `70` |
+| Tile ID | `20` |
+| Stages ID | `37` |
+| Split-K | `1` |
+| Reduction scheme | `0` |
+| Custom option / CTA swizzle | `0 / 0` |
+| Required workspace | `0 bytes` |
+| Supplied workspace | `64 MiB` sweep; `0 bytes` block/trajectory gates |
+
+The baseline heuristic's negative split-K value is an undocumented library
+sentinel. CUDA 13 has no documented public attribute that directly selects
+Stream-K, so the sentinel itself is not interpreted as a public Stream-K
+control. The launched baseline symbol does end in `_stream_k`.
+
+## Canonical Operation
+
+- Logical GEMM: `M=37,810, N=5,376, K=14,336`.
+- Descriptor GEMM after accepted producer padding: `M=37,824`.
+- Inputs: packed NVFP4 E2M1 activation and weight with vector-16 UE4M3 scales.
+- Accumulation/scalars: FP32; output: BF16.
+- Accepted activation producer: `vortex_native_quantize_swiglu_nvfp4`.
+- Image: `sha256:1d340e14cb6fc45ccfdbe63dde8db2a2b3aea94b493702c8a08e1f8d5b4f7b83`.
+
+## Timing
+
+The 20-round isolated sweep alternated AB/BA order in one loaded model.
+
+| Measurement | Baseline | Split-K-1 | Result |
+| --- | ---: | ---: | ---: |
+| FC2 p50 | `53.618 ms` | `15.636 ms` | `3.43x` |
+| Dense throughput p50 | `108.70 TFLOP/s` | `372.74 TFLOP/s` | `3.43x` |
+| Producer + FC2 p50 | `76.737 ms` | `38.823 ms` | `49.41%` faster |
+
+The negative-sentinel candidate reproduced the production-equivalent baseline
+within noise. Explicit split-K factors `2`, `4`, `8`, and `16` were rejected by
+`cublasLtMatmulAlgoCheck` for the tested configuration space.
+
+Complete 20-round block gates replaced only
+`block.mlp.fc2.forward_swiglu`:
+
+| Block | Baseline p50 | Candidate p50 | Improvement |
+| ---: | ---: | ---: | ---: |
+| 0 | `457.891 ms` | `420.536 ms` | `8.16%` |
+| 24 | `459.928 ms` | `420.243 ms` | `8.63%` |
+| 49 | `458.054 ms` | `417.364 ms` | `8.88%` |
+
+All block outputs were byte-exact against both the paired baseline and retained
+traversal.
+
+| Trajectory | Baseline | Candidate | Improvement | Correctness |
+| --- | ---: | ---: | ---: | --- |
+| Two-step | `45.962 s` | `42.516 s` | `7.50%` | video/audio BF16 exact |
+| Canonical 12-step | `278.201 s` | `255.371 s` | `8.21%` | video/audio BF16 exact |
+
+Trajectory values are single paired runs in baseline-then-candidate
+order, not repeated medians. Their approximately `38 ms` per-FC2 savings agree
+with the alternating isolated and complete-block gates, but the precise
+end-to-end percentages retain run-order uncertainty.
+
+The 12-step saving of `22.829 s` matches approximately 600 FC2 invocations
+multiplied by the isolated roughly `38 ms` saving.
+
+## Traffic Attribution
+
+The one-pass distinct-data footprint is:
+
+| Category | Bytes |
+| --- | ---: |
+| Activation QDATA + scales | `305,012,736` |
+| Weight QDATA + scales | `43,352,064` |
+| Padded BF16 output | `406,683,648` |
+| Total | `755,048,448` |
+
+Both schedules issue the same nominal operand requests:
+
+| Requested category | Bytes | Geometric reuse |
+| --- | ---: | ---: |
+| Weight QDATA + scales | `12,832,210,944` | `296x` |
+| Activation QDATA + scales | `12,832,210,944` | about `42.1x` |
+| Total L2 operand reads | `25,664,424,960` | unchanged |
+
+The schedule does not eliminate CTA-level rereading. It changes whether those
+rereads remain cache-resident:
+
+| NCU metric | Baseline `_stream_k` | Split-K-1 |
+| --- | ---: | ---: |
+| Main-kernel duration | `55.057 ms` | `17.031 ms` |
+| Combined L2 hit rate | `53.32%` | `91.10%` |
+| L2 read-miss bytes | `11.762 GB` | `1.909 GB` |
+| L2 miss-byte proxy | `12.169 GB` | `2.316 GB` |
+| Sysmem traffic proxy | `12.176 GB` | `2.327 GB` |
+| Off-chip amplification | `16.12x` | `3.07x` |
+| SM throughput | `25.11%` | `81.79%` |
+| Tensor-pipe active share | `25.01%` | `82.42%` |
+| Eligible warps/scheduler | `0.073` | `0.220` |
+| Issue rate | `0.057` | `0.168` |
+
+GB10 does not expose the usual discrete-GPU DRAM byte counters in these
+captures. L2 misses and the reported sysmem fill/write sectors are used as the
+off-chip proxy.
+
+The `9.853 GB` reduction in L2 read misses is avoided operand rereading. NCU
+aggregates the two TMA input descriptors, so it cannot defensibly assign exact
+miss-byte totals separately to weights and activations. The request geometry
+does prove that each input family accounts for `12.832 GB` of requested reads.
+
+There is no material global partial-accumulator or output-reduction traffic:
+
+- Candidate workspace is zero.
+- Neither kernel issues global atomics or global reduction operations.
+- No auxiliary reduction kernel is launched.
+- Candidate writes exactly one padded BF16 output; baseline writes only
+  `73,728` bytes more.
+- Baseline-only local stack traffic is about `22.35 MB` at L1 and almost none
+  reaches sysmem.
+
+The baseline's dominant scheduler stall is sleeping. Source-correlated samples
+land in `NANOSLEEP.SYNCS` polling around synchronization phase checks. Launch
+resources are otherwise the same: 12,432 CTAs, 384 threads/CTA, 168 registers
+per thread, 89,088 bytes allocated shared memory, and 25% theoretical
+occupancy. The gain therefore comes from traversal locality and reduced
+synchronization waiting, not occupancy or a reduction workspace.
+
+## Evidence
+
+- `benchmarks/gb10-fc2-nvfp4-library-sweep-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-block-gate-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-trajectory-2step-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-trajectory-12step-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-baseline-profile-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-splitk1-profile-20260825.json`
+- `benchmarks/gb10-fc2-nvfp4-baseline-20260825.csv`
+- `benchmarks/gb10-fc2-nvfp4-splitk1-20260825.csv`
+- `benchmarks/gb10-fc2-nvfp4-baseline-20260825.ncu-rep`
+- `benchmarks/gb10-fc2-nvfp4-splitk1-20260825.ncu-rep`
+
+Production source and configuration were not changed by this study. The
+candidate is validated only for the canonical descriptor shape and current
+GB10/CUDA-library combination. Promotion still requires shape-specific
+`AlgoCheck` with a safe fallback, integration behind the existing
+`Nvfp4Linear` boundary, broader shape tests, and deployment validation.
diff --git a/research/fc2_nvfp4_scheduling/__init__.py b/research/fc2_nvfp4_scheduling/__init__.py
new file mode 100644
index 0000000..cba87aa
--- /dev/null
+++ b/research/fc2_nvfp4_scheduling/__init__.py
@@ -0,0 +1 @@
+"""Isolated cuBLASLt NVFP4 FC2 scheduling research."""
diff --git a/research/fc2_nvfp4_scheduling/candidate_splitk1.json b/research/fc2_nvfp4_scheduling/candidate_splitk1.json
new file mode 100644
index 0000000..e1ce3ec
--- /dev/null
+++ b/research/fc2_nvfp4_scheduling/candidate_splitk1.json
@@ -0,0 +1,9 @@
+{
+  "algorithm_id": 70,
+  "tile_id": 20,
+  "stages_id": 37,
+  "split_k": 1,
+  "reduction_scheme": 0,
+  "custom_option": 0,
+  "cta_swizzle": 0
+}
diff --git a/research/fc2_nvfp4_scheduling/fc2_nvfp4_lt.cpp b/research/fc2_nvfp4_scheduling/fc2_nvfp4_lt.cpp
new file mode 100644
index 0000000..3cf79d8
--- /dev/null
+++ b/research/fc2_nvfp4_scheduling/fc2_nvfp4_lt.cpp
@@ -0,0 +1,388 @@
+#include <torch/extension.h>
+
+#include <ATen/cuda/CUDAContext.h>
+#include <c10/cuda/CUDAGuard.h>
+#include <cublasLt.h>
+#include <cuda_runtime_api.h>
+
+#include <array>
+#include <cstdint>
+#include <stdexcept>
+#include <string>
+#include <vector>
+
+namespace py = pybind11;
+
+#define LT_CHECK(call)                                                        \
+  do {                                                                        \
+    const cublasStatus_t status_ = (call);                                     \
+    if (status_ != CUBLAS_STATUS_SUCCESS)                                      \
+      throw std::runtime_error(std::string(#call) + " failed: " +             \
+                               std::to_string(static_cast<int>(status_)));     \
+  } while (0)
+
+namespace {
+
+thread_local cublasLtHandle_t handle = nullptr;
+
+cublasLtHandle_t get_handle() {
+  if (!handle) LT_CHECK(cublasLtCreate(&handle));
+  return handle;
+}
+
+void require_cuda_contiguous(const torch::Tensor& tensor, const char* name) {
+  TORCH_CHECK(tensor.is_cuda(), name, " must be CUDA");
+  TORCH_CHECK(tensor.is_contiguous(), name, " must be contiguous");
+}
+
+void require_same_device(const torch::Tensor& tensor,
+                         const torch::Tensor& reference,
+                         const char* name) {
+  TORCH_CHECK(tensor.device() == reference.device(), name,
+              " must be on the same CUDA device as activation_qdata");
+}
+
+struct Problem {
+  cublasLtMatmulDesc_t operation = nullptr;
+  cublasLtMatrixLayout_t a = nullptr, b = nullptr, c = nullptr, d = nullptr;
+
+  Problem(const torch::Tensor& activation_qdata,
+          const torch::Tensor& activation_block_scale,
+          const torch::Tensor& weight_qdata,
+          const torch::Tensor& weight_block_scale) {
+    require_cuda_contiguous(activation_qdata, "activation_qdata");
+    require_cuda_contiguous(activation_block_scale, "activation_block_scale");
+    require_cuda_contiguous(weight_qdata, "weight_qdata");
+    require_cuda_contiguous(weight_block_scale, "weight_block_scale");
+    require_same_device(activation_block_scale, activation_qdata,
+                        "activation_block_scale");
+    require_same_device(weight_qdata, activation_qdata, "weight_qdata");
+    require_same_device(weight_block_scale, activation_qdata,
+                        "weight_block_scale");
+    TORCH_CHECK(activation_qdata.scalar_type() == at::kByte &&
+                weight_qdata.scalar_type() == at::kByte,
+                "packed NVFP4 operands must use uint8 storage");
+    TORCH_CHECK(activation_qdata.dim() == 2 && weight_qdata.dim() == 2,
+                "packed NVFP4 operands must be rank two");
+    TORCH_CHECK(activation_block_scale.element_size() == 1 &&
+                weight_block_scale.element_size() == 1,
+                "NVFP4 block scales must use one-byte E4M3 storage");
+    TORCH_CHECK(activation_qdata.size(1) == weight_qdata.size(1),
+                "packed K dimensions differ");
+
+    // This is deliberately the same column-major reinterpretation used by
+    // Comfy Kitchen 0.2.31: weight is Lt A, activation is Lt B, and D is D^T.
+    const int64_t m = weight_qdata.size(0);       // row-major N
+    const int64_t n = activation_qdata.size(0);  // row-major padded M
+    const int64_t k = activation_qdata.size(1) * 2;
+    TORCH_CHECK(activation_block_scale.numel() >= n * (k / 16),
+                "activation_block_scale storage is too small");
+    TORCH_CHECK(weight_block_scale.numel() >= m * (k / 16),
+                "weight_block_scale storage is too small");
+    LT_CHECK(cublasLtMatmulDescCreate(&operation, CUBLAS_COMPUTE_32F,
+                                     CUDA_R_32F));
+    cublasLtMatmulMatrixScale_t scale_mode =
+        CUBLASLT_MATMUL_MATRIX_SCALE_VEC16_UE4M3;
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_A_SCALE_MODE, &scale_mode,
+        sizeof(scale_mode)));
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_B_SCALE_MODE, &scale_mode,
+        sizeof(scale_mode)));
+    const cublasOperation_t transa = CUBLAS_OP_T;
+    const cublasOperation_t transb = CUBLAS_OP_N;
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_TRANSA, &transa, sizeof(transa)));
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_TRANSB, &transb, sizeof(transb)));
+    const void* a_scale = weight_block_scale.data_ptr();
+    const void* b_scale = activation_block_scale.data_ptr();
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_A_SCALE_POINTER, &a_scale,
+        sizeof(a_scale)));
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_B_SCALE_POINTER, &b_scale,
+        sizeof(b_scale)));
+    const cublasDataType_t scale_type = CUDA_R_32F;
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_SCALE_TYPE, &scale_type,
+        sizeof(scale_type)));
+    const cublasLtPointerMode_t pointer_mode = CUBLASLT_POINTER_MODE_DEVICE;
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_POINTER_MODE, &pointer_mode,
+        sizeof(pointer_mode)));
+    const cublasLtEpilogue_t epilogue = CUBLASLT_EPILOGUE_DEFAULT;
+    LT_CHECK(cublasLtMatmulDescSetAttribute(
+        operation, CUBLASLT_MATMUL_DESC_EPILOGUE, &epilogue,
+        sizeof(epilogue)));
+
+    LT_CHECK(cublasLtMatrixLayoutCreate(&a, CUDA_R_4F_E2M1, k, m, k));
+    LT_CHECK(cublasLtMatrixLayoutCreate(&b, CUDA_R_4F_E2M1, k, n, k));
+    LT_CHECK(cublasLtMatrixLayoutCreate(&c, CUDA_R_16BF, m, n, m));
+    LT_CHECK(cublasLtMatrixLayoutCreate(&d, CUDA_R_16BF, m, n, m));
+  }
+
+  ~Problem() {
+    if (d) cublasLtMatrixLayoutDestroy(d);
+    if (c) cublasLtMatrixLayoutDestroy(c);
+    if (b) cublasLtMatrixLayoutDestroy(b);
+    if (a) cublasLtMatrixLayoutDestroy(a);
+    if (operation) cublasLtMatmulDescDestroy(operation);
+  }
+};
+
+template <typename T>
+bool config_get(const cublasLtMatmulAlgo_t& algo,
+                cublasLtMatmulAlgoConfigAttributes_t attr, T* value) {
+  size_t written = 0;
+  return cublasLtMatmulAlgoConfigGetAttribute(&algo, attr, value,
+                                               sizeof(T), &written) ==
+             CUBLAS_STATUS_SUCCESS &&
+         written == sizeof(T);
+}
+
+template <typename T>
+void put_config(py::dict& result, const char* name,
+                const cublasLtMatmulAlgo_t& algo,
+                cublasLtMatmulAlgoConfigAttributes_t attr) {
+  T value{};
+  if (config_get(algo, attr, &value)) result[name] = value;
+}
+
+template <typename T>
+void put_cap_scalar(py::dict& caps, const char* name,
+                    const cublasLtMatmulAlgo_t& algo,
+                    cublasLtMatmulAlgoCapAttributes_t attr) {
+  T value{};
+  size_t written = 0;
+  if (cublasLtMatmulAlgoCapGetAttribute(&algo, attr, &value, sizeof(value),
+                                        &written) == CUBLAS_STATUS_SUCCESS &&
+      written == sizeof(value))
+    caps[name] = value;
+}
+
+void put_cap_array(py::dict& caps, const char* name,
+                   const cublasLtMatmulAlgo_t& algo,
+                   cublasLtMatmulAlgoCapAttributes_t attr) {
+  size_t bytes = 0;
+  if (cublasLtMatmulAlgoCapGetAttribute(&algo, attr, nullptr, 0, &bytes) !=
+          CUBLAS_STATUS_SUCCESS ||
+      bytes == 0)
+    return;
+  std::vector<uint32_t> values((bytes + sizeof(uint32_t) - 1) /
+                               sizeof(uint32_t));
+  size_t written = 0;
+  if (cublasLtMatmulAlgoCapGetAttribute(&algo, attr, values.data(), bytes,
+                                        &written) != CUBLAS_STATUS_SUCCESS)
+    return;
+  values.resize(written / sizeof(uint32_t));
+  caps[name] = values;
+}
+
+py::dict describe(const cublasLtMatmulAlgo_t& algo,
+                  const cublasLtMatmulHeuristicResult_t& checked,
+                  cublasStatus_t api_status) {
+  py::dict result;
+  for (const char* name : {"algorithm_id", "tile_id", "stages_id", "split_k",
+                           "reduction_scheme", "custom_option", "cta_swizzle",
+                           "inner_shape", "cluster_shape"})
+    result[name] = py::none();
+  put_config<int>(result, "algorithm_id", algo, CUBLASLT_ALGO_CONFIG_ID);
+  put_config<uint32_t>(result, "tile_id", algo, CUBLASLT_ALGO_CONFIG_TILE_ID);
+  put_config<uint32_t>(result, "stages_id", algo,
+                       CUBLASLT_ALGO_CONFIG_STAGES_ID);
+  // CUDA 13 documents SPLITK_NUM as int32_t. Preserve negative library
+  // sentinel values instead of wrapping them into fictitious huge factors.
+  put_config<int32_t>(result, "split_k", algo,
+                      CUBLASLT_ALGO_CONFIG_SPLITK_NUM);
+  put_config<uint32_t>(result, "reduction_scheme", algo,
+                       CUBLASLT_ALGO_CONFIG_REDUCTION_SCHEME);
+  put_config<uint32_t>(result, "custom_option", algo,
+                       CUBLASLT_ALGO_CONFIG_CUSTOM_OPTION);
+  put_config<uint32_t>(result, "cta_swizzle", algo,
+                       CUBLASLT_ALGO_CONFIG_CTA_SWIZZLING);
+#if CUDA_VERSION >= 12000
+  put_config<uint32_t>(result, "inner_shape", algo,
+                       CUBLASLT_ALGO_CONFIG_INNER_SHAPE_ID);
+  put_config<uint32_t>(result, "cluster_shape", algo,
+                       CUBLASLT_ALGO_CONFIG_CLUSTER_SHAPE_ID);
+#endif
+  result["required_workspace_bytes"] = checked.workspaceSize;
+  result["waves"] = checked.wavesCount;
+  result["state"] = static_cast<int>(checked.state);
+  result["api_status"] = static_cast<int>(api_status);
+  result["valid"] = api_status == CUBLAS_STATUS_SUCCESS &&
+                    checked.state == CUBLAS_STATUS_SUCCESS;
+
+  py::dict caps;
+  put_cap_scalar<int>(caps, "split_k_support", algo,
+                      CUBLASLT_ALGO_CAP_SPLITK_SUPPORT);
+  put_cap_scalar<uint32_t>(caps, "reduction_scheme_mask", algo,
+                           CUBLASLT_ALGO_CAP_REDUCTION_SCHEME_MASK);
+  put_cap_scalar<uint32_t>(caps, "cta_swizzle_support", algo,
+                           CUBLASLT_ALGO_CAP_CTA_SWIZZLING_SUPPORT);
+  put_cap_scalar<int>(caps, "custom_option_max", algo,
+                      CUBLASLT_ALGO_CAP_CUSTOM_OPTION_MAX);
+  put_cap_scalar<int>(caps, "strided_batch_support", algo,
+                      CUBLASLT_ALGO_CAP_STRIDED_BATCH_SUPPORT);
+  put_cap_scalar<int>(caps, "out_of_place_result_support", algo,
+                      CUBLASLT_ALGO_CAP_OUT_OF_PLACE_RESULT_SUPPORT);
+  put_cap_array(caps, "tile_ids", algo, CUBLASLT_ALGO_CAP_TILE_IDS);
+  put_cap_array(caps, "stages_ids", algo, CUBLASLT_ALGO_CAP_STAGES_IDS);
+  caps["inner_cluster_shape_capability_note"] =
+      "This CUDA 13 cublasLt.h exposes config IDs but no public capability "
+      "attributes that enumerate inner/cluster shape IDs.";
+  result["capabilities"] = caps;
+  return result;
+}
+
+cublasLtMatmulAlgo_t init_algo(int algorithm_id) {
+  cublasLtMatmulAlgo_t algo{};
+  LT_CHECK(cublasLtMatmulAlgoInit(
+      get_handle(), CUBLAS_COMPUTE_32F, CUDA_R_32F, CUDA_R_4F_E2M1,
+      CUDA_R_4F_E2M1, CUDA_R_16BF, CUDA_R_16BF, algorithm_id, &algo));
+  return algo;
+}
+
+template <typename T>
+void maybe_set(cublasLtMatmulAlgo_t* algo, const py::dict& config,
+               const char* key, cublasLtMatmulAlgoConfigAttributes_t attr) {
+  if (!config.contains(key) || config[key].is_none()) return;
+  const T value = config[key].cast<T>();
+  LT_CHECK(cublasLtMatmulAlgoConfigSetAttribute(algo, attr, &value,
+                                                sizeof(value)));
+}
+
+cublasLtMatmulAlgo_t configured_algo(const py::dict& config) {
+  TORCH_CHECK(config.contains("algorithm_id"), "algorithm_id is required");
+  auto algo = init_algo(config["algorithm_id"].cast<int>());
+  maybe_set<uint32_t>(&algo, config, "tile_id", CUBLASLT_ALGO_CONFIG_TILE_ID);
+  maybe_set<uint32_t>(&algo, config, "stages_id",
+                      CUBLASLT_ALGO_CONFIG_STAGES_ID);
+  maybe_set<int32_t>(&algo, config, "split_k",
+                     CUBLASLT_ALGO_CONFIG_SPLITK_NUM);
+  maybe_set<uint32_t>(&algo, config, "reduction_scheme",
+                      CUBLASLT_ALGO_CONFIG_REDUCTION_SCHEME);
+  maybe_set<uint32_t>(&algo, config, "custom_option",
+                      CUBLASLT_ALGO_CONFIG_CUSTOM_OPTION);
+  maybe_set<uint32_t>(&algo, config, "cta_swizzle",
+                      CUBLASLT_ALGO_CONFIG_CTA_SWIZZLING);
+#if CUDA_VERSION >= 12000
+  maybe_set<uint32_t>(&algo, config, "inner_shape",
+                      CUBLASLT_ALGO_CONFIG_INNER_SHAPE_ID);
+  maybe_set<uint32_t>(&algo, config, "cluster_shape",
+                      CUBLASLT_ALGO_CONFIG_CLUSTER_SHAPE_ID);
+#endif
+  return algo;
+}
+
+py::list enumerate(torch::Tensor activation_qdata,
+                   torch::Tensor activation_block_scale,
+                   torch::Tensor weight_qdata,
+                   torch::Tensor weight_block_scale,
+                   int64_t max_workspace, int requested_count) {
+  c10::cuda::CUDAGuard guard(activation_qdata.device());
+  Problem problem(activation_qdata, activation_block_scale, weight_qdata,
+                  weight_block_scale);
+  cublasLtMatmulPreference_t preference = nullptr;
+  LT_CHECK(cublasLtMatmulPreferenceCreate(&preference));
+  LT_CHECK(cublasLtMatmulPreferenceSetAttribute(
+      preference, CUBLASLT_MATMUL_PREF_MAX_WORKSPACE_BYTES, &max_workspace,
+      sizeof(max_workspace)));
+  std::vector<cublasLtMatmulHeuristicResult_t> found(requested_count);
+  int returned = 0;
+  const auto status = cublasLtMatmulAlgoGetHeuristic(
+      get_handle(), problem.operation, problem.a, problem.b, problem.c,
+      problem.d, preference, requested_count, found.data(), &returned);
+  cublasLtMatmulPreferenceDestroy(preference);
+  LT_CHECK(status);
+  py::list output;
+  for (int i = 0; i < returned; ++i)
+    output.append(describe(found[i].algo, found[i], CUBLAS_STATUS_SUCCESS));
+  return output;
+}
+
+py::dict check(torch::Tensor activation_qdata,
+               torch::Tensor activation_block_scale,
+               torch::Tensor weight_qdata,
+               torch::Tensor weight_block_scale, py::dict config) {
+  c10::cuda::CUDAGuard guard(activation_qdata.device());
+  Problem problem(activation_qdata, activation_block_scale, weight_qdata,
+                  weight_block_scale);
+  auto algo = configured_algo(config);
+  cublasLtMatmulHeuristicResult_t result{};
+  const auto status = cublasLtMatmulAlgoCheck(
+      get_handle(), problem.operation, problem.a, problem.b, problem.c,
+      problem.d, &algo, &result);
+  return describe(algo, result, status);
+}
+
+void run(torch::Tensor activation_qdata,
+         torch::Tensor activation_block_scale,
+         torch::Tensor weight_qdata, torch::Tensor weight_block_scale,
+         torch::Tensor alpha, torch::Tensor beta, torch::Tensor output,
+         torch::Tensor workspace, py::dict config) {
+  c10::cuda::CUDAGuard guard(activation_qdata.device());
+  require_cuda_contiguous(alpha, "alpha");
+  require_cuda_contiguous(beta, "beta");
+  require_cuda_contiguous(output, "output");
+  require_cuda_contiguous(workspace, "workspace");
+  require_same_device(alpha, activation_qdata, "alpha");
+  require_same_device(beta, activation_qdata, "beta");
+  require_same_device(output, activation_qdata, "output");
+  require_same_device(workspace, activation_qdata, "workspace");
+  TORCH_CHECK(alpha.scalar_type() == at::kFloat && alpha.numel() == 1,
+              "alpha must be one device FP32 value");
+  TORCH_CHECK(beta.scalar_type() == at::kFloat && beta.numel() == 1,
+              "beta must be one device FP32 value");
+  TORCH_CHECK(output.scalar_type() == at::kBFloat16 && output.dim() == 2,
+              "output must be rank-two BF16");
+  TORCH_CHECK(workspace.scalar_type() == at::kByte,
+              "workspace must use uint8 storage");
+  TORCH_CHECK(output.size(0) == activation_qdata.size(0) &&
+              output.size(1) == weight_qdata.size(0),
+              "output must be [packed activation rows, weight rows]");
+  Problem problem(activation_qdata, activation_block_scale, weight_qdata,
+                  weight_block_scale);
+  auto algo = configured_algo(config);
+  cublasLtMatmulHeuristicResult_t checked{};
+  LT_CHECK(cublasLtMatmulAlgoCheck(get_handle(), problem.operation, problem.a,
+                                  problem.b, problem.c, problem.d, &algo,
+                                  &checked));
+  TORCH_CHECK(checked.state == CUBLAS_STATUS_SUCCESS,
+              "selected algorithm failed AlgoCheck with state ",
+              static_cast<int>(checked.state));
+  TORCH_CHECK(checked.workspaceSize <= static_cast<size_t>(workspace.numel()),
+              "selected algorithm requires ", checked.workspaceSize,
+              " workspace bytes but caller supplied ", workspace.numel());
+  const auto stream = at::cuda::getCurrentCUDAStream(
+      activation_qdata.get_device()).stream();
+  void* workspace_ptr = workspace.numel() ? workspace.data_ptr() : nullptr;
+  LT_CHECK(cublasLtMatmul(
+      get_handle(), problem.operation, alpha.data_ptr(), weight_qdata.data_ptr(),
+      problem.a, activation_qdata.data_ptr(), problem.b, beta.data_ptr(),
+      output.data_ptr(), problem.c, output.data_ptr(), problem.d, &algo,
+      workspace_ptr, workspace.numel(), stream));
+}
+
+py::dict build_info() {
+  py::dict result;
+  result["cuda_version"] = CUDA_VERSION;
+  result["cublas_version"] = CUBLAS_VERSION;
+  result["stream_k_public_control"] = false;
+  result["stream_k_note"] =
+      "CUDA 13 cuBLASLt exposes no documented MatmulAlgoConfig attribute "
+      "that directly selects Stream-K. Negative SPLITK_NUM values returned "
+      "by heuristics are preserved as undocumented library sentinels, not "
+      "claimed as public Stream-K control.";
+  return result;
+}
+
+}  // namespace
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, module) {
+  module.def("enumerate", &enumerate);
+  module.def("check", &check);
+  module.def("run", &run);
+  module.def("build_info", &build_info);
+}
diff --git a/tools/benchmark_fc2_nvfp4_algorithms.py b/tools/benchmark_fc2_nvfp4_algorithms.py
new file mode 100644
index 0000000..49de326
--- /dev/null
+++ b/tools/benchmark_fc2_nvfp4_algorithms.py
@@ -0,0 +1,729 @@
+"""Characterize isolated cuBLASLt NVFP4 scheduling at the real H3 FC2 boundary."""
+
+from __future__ import annotations
+
+import argparse
+import hashlib
+import json
+import os
+import platform
+import statistics
+import subprocess
+import sys
+import time
+import types
+from pathlib import Path
+from typing import Any, Callable
+
+import torch
+
+from h3_blackwell_runtime.checkpoint import H3Checkpoint
+from h3_blackwell_runtime.denoiser import H3PackedDenoiser
+from h3_blackwell_runtime.nvfp4_quant import vortex_native_quantize_swiglu_nvfp4
+from h3_blackwell_runtime.packing import H3PromptPacker
+from h3_blackwell_runtime.rope import h3_rope_rotation
+from h3_blackwell_runtime.sampler import _audio_sigma, _model_sigma, beta_sigmas, sample_video_res_multistep
+from h3_blackwell_runtime.t2v import random_av_latents
+
+
+ROOT = Path(__file__).resolve().parents[1]
+SOURCE = ROOT / "research" / "fc2_nvfp4_scheduling" / "fc2_nvfp4_lt.cpp"
+DEFAULT_BUDGETS = (0, 4 << 20, 8 << 20, 16 << 20, 32 << 20, 64 << 20)
+DEFAULT_BLOCKS = (0, 24, 49)
+ENV_PREFIXES = ("H3_", "COMFY_KITCHEN_", "CUDA_", "TORCH_")
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--mode", choices=("compile", "characterize", "sweep", "selected", "profile", "block-gate", "trajectory"), default="characterize")
+    parser.add_argument("--model-path", default="/models/minimax_h3_fl2va_pruned_nvfp4.safetensors")
+    parser.add_argument("--capture", type=Path, help="Directory containing input.pt, or an input .pt file")
+    parser.add_argument("--output", type=Path, default=Path("/output/h3-blackwell-runtime/benchmarks/fc2-nvfp4-scheduling.json"))
+    parser.add_argument("--build-directory", type=Path)
+    parser.add_argument("--verbose-build", action="store_true")
+    parser.add_argument("--width", type=int, default=1344)
+    parser.add_argument("--height", type=int, default=768)
+    parser.add_argument("--frames", type=int, default=124)
+    parser.add_argument("--steps", type=int, default=12)
+    parser.add_argument("--sampler-step", type=int, default=1)
+    parser.add_argument("--seed", type=int, default=440420)
+    parser.add_argument("--text-tokens", type=int, default=100)
+    parser.add_argument("--blocks", type=int, nargs="+", default=list(DEFAULT_BLOCKS))
+    parser.add_argument("--probe-block", type=int, default=24)
+    parser.add_argument("--workspace-budgets", type=int, nargs="+", default=list(DEFAULT_BUDGETS))
+    parser.add_argument("--requested-count", type=int, default=32)
+    parser.add_argument("--explicit-split-k", type=int, nargs="+", default=[1, 2, 4, 8, 16])
+    parser.add_argument("--max-explicit-checks", type=int, default=64)
+    parser.add_argument("--candidate-limit", type=int, default=8)
+    parser.add_argument("--candidate", help="JSON object, JSON file, or enumerate:<budget-index>:<result-index>")
+    parser.add_argument("--profile-target", choices=("baseline", "candidate"), default="candidate")
+    parser.add_argument("--workspace-bytes", type=int, default=32 << 20)
+    parser.add_argument("--warmup", type=int, default=2)
+    parser.add_argument("--rounds", type=int, default=6)
+    parser.add_argument("--checkpoint-sha256", action="store_true", help="Expensive: actually calculate and record the checkpoint SHA-256")
+    return parser.parse_args()
+
+
+def load_extension(args: argparse.Namespace):
+    from torch.utils.cpp_extension import CUDA_HOME, load
+
+    if CUDA_HOME is None:
+        raise RuntimeError("CUDA_HOME is unavailable; CUDA 12.9 or newer headers are required")
+    kwargs: dict[str, Any] = {}
+    if args.build_directory is not None:
+        args.build_directory.mkdir(parents=True, exist_ok=True)
+        kwargs["build_directory"] = str(args.build_directory)
+    return load(
+        name="h3_fc2_nvfp4_lt_schedule",
+        sources=[str(SOURCE)],
+        extra_include_paths=[str(Path(CUDA_HOME) / "include")],
+        extra_cflags=["-O2", "-std=c++17"],
+        extra_ldflags=["-L" + str(Path(CUDA_HOME) / "lib64"), "-lcublasLt", "-lcublas", "-lcudart"],
+        verbose=args.verbose_build,
+        **kwargs,
+    )
+
+
+def sha256_file(path: Path) -> str:
+    digest = hashlib.sha256()
+    with path.open("rb") as handle:
+        for chunk in iter(lambda: handle.read(8 << 20), b""):
+            digest.update(chunk)
+    return digest.hexdigest()
+
+
+def tensor_sha256(tensor: torch.Tensor) -> str:
+    immutable = tensor.detach().contiguous().clone().view(torch.uint16).cpu()
+    return hashlib.sha256(immutable.numpy().tobytes()).hexdigest()
+
+
+def compare(actual: torch.Tensor, expected: torch.Tensor) -> dict[str, Any]:
+    actual_copy = actual.detach().clone()
+    expected_copy = expected.detach().clone()
+    different = int(torch.count_nonzero(actual_copy != expected_copy))
+    delta = (actual_copy.float() - expected_copy.float()).abs()
+    return {
+        "bf16_exact": different == 0,
+        "different_elements": different,
+        "max_abs": float(delta.max()),
+        "mean_abs": float(delta.mean()),
+        "actual_sha256": tensor_sha256(actual_copy),
+        "expected_sha256": tensor_sha256(expected_copy),
+    }
+
+
+def percentile(values: list[float], fraction: float) -> float:
+    ordered = sorted(values)
+    if len(ordered) == 1:
+        return ordered[0]
+    position = (len(ordered) - 1) * fraction
+    lower = int(position)
+    weight = position - lower
+    return ordered[lower] * (1.0 - weight) + ordered[min(lower + 1, len(ordered) - 1)] * weight
+
+
+def timing_summary(milliseconds: list[float], m: int | None = None, n: int | None = None, k: int | None = None) -> dict[str, Any]:
+    result: dict[str, Any] = {
+        "samples_ms": milliseconds,
+        "p50_ms": percentile(milliseconds, 0.50),
+        "p95_ms": percentile(milliseconds, 0.95),
+        "mean_ms": statistics.fmean(milliseconds),
+    }
+    if m is not None and n is not None and k is not None:
+        result["dense_tflop_s_p50"] = 2.0 * m * n * k / (result["p50_ms"] * 1.0e9)
+    return result
+
+
+def timed_cuda(call: Callable[[], torch.Tensor]) -> tuple[float, torch.Tensor]:
+    start = torch.cuda.Event(enable_timing=True)
+    end = torch.cuda.Event(enable_timing=True)
+    start.record()
+    output = call()
+    end.record()
+    end.synchronize()
+    return float(start.elapsed_time(end)), output
+
+
+def environment(args: argparse.Namespace, extension) -> dict[str, Any]:
+    checkpoint = Path(args.model_path)
+    try:
+        commit = subprocess.run(
+            ["git", "rev-parse", "HEAD"], cwd=ROOT, check=True,
+            capture_output=True, text=True,
+        ).stdout.strip()
+    except (OSError, subprocess.CalledProcessError):
+        commit = None
+    result = {
+        "platform": platform.platform(),
+        "python": sys.version,
+        "torch": torch.__version__,
+        "torch_cuda": torch.version.cuda,
+        "device": torch.cuda.get_device_name(),
+        "device_capability": list(torch.cuda.get_device_capability()),
+        "driver": torch.cuda.driver_version() if hasattr(torch.cuda, "driver_version") else None,
+        "git_commit": commit,
+        "checkpoint_path": str(checkpoint),
+        "checkpoint_sha256": sha256_file(checkpoint) if args.checkpoint_sha256 else None,
+        "checkpoint_hash_note": "calculated" if args.checkpoint_sha256 else "not calculated",
+        "environment_switches": {key: value for key, value in sorted(os.environ.items()) if key.startswith(ENV_PREFIXES)},
+        "extension": dict(extension.build_info()),
+    }
+    try:
+        import comfy_kitchen
+        result["comfy_kitchen"] = getattr(comfy_kitchen, "__version__", "0.2.31 package without __version__")
+    except Exception as error:
+        result["comfy_kitchen"] = f"import error: {error}"
+    return result
+
+
+def make_workload(args: argparse.Namespace, checkpoint: H3Checkpoint, model: H3PackedDenoiser):
+    torch.manual_seed(args.seed)
+    if args.capture is not None:
+        path = args.capture / "input.pt" if args.capture.is_dir() else args.capture
+        payload = torch.load(path, map_location="cuda", weights_only=False)
+        hidden = payload["hidden"].to("cuda").contiguous()
+        timesteps = payload["timesteps"].to("cuda")
+        positions = payload["position_ids"].to("cuda")
+        segments = payload["segments"]
+        metadata = {"capture": str(path)}
+    else:
+        packer = H3PromptPacker(checkpoint)
+        video, audio, aligned_frames = random_av_latents(
+            args.width, args.height, args.frames, args.seed, device="cuda",
+        )
+        sigma = beta_sigmas(args.steps, device="cuda")[args.sampler_step - 1]
+        native_audio = audio.to(torch.bfloat16) * (_audio_sigma(sigma) / sigma)
+        text = torch.randn(1, args.text_tokens, 5376, device="cuda", dtype=torch.bfloat16)
+        hidden, timesteps, segments, positions, _, _ = packer(
+            text, video, native_audio, _model_sigma(sigma),
+        )
+        metadata = {
+            "resolution": [args.width, args.height],
+            "frames": aligned_frames,
+            "steps": args.steps,
+            "sampler_step": args.sampler_step,
+            "seed": args.seed,
+            "text_tokens": args.text_tokens,
+        }
+        canonical = (args.width, args.height, args.frames, args.seed, args.sampler_step, args.text_tokens) == (1344, 768, 124, 440420, 1, 100)
+        if canonical and hidden.shape[0] != 37_810:
+            raise RuntimeError(f"canonical workload must contain 37,810 tokens, got {hidden.shape[0]}")
+    rotation = h3_rope_rotation(positions.to("cuda"), model.backbone.inv_freq, torch.bfloat16)
+    metadata.update({"tokens": hidden.shape[0], "hidden_shape": list(hidden.shape), "segments": segments})
+    return hidden, timesteps, rotation, segments, metadata
+
+
+def capture_boundaries(args: argparse.Namespace, model: H3PackedDenoiser, hidden, timesteps, rotation, segments):
+    wanted = set(args.blocks)
+    if wanted != set(DEFAULT_BLOCKS):
+        missing = set(DEFAULT_BLOCKS) - wanted
+        if missing:
+            raise ValueError(f"--blocks must retain required blocks 0,24,49; missing {sorted(missing)}")
+    block_inputs: dict[int, torch.Tensor] = {}
+    block_outputs: dict[int, torch.Tensor] = {}
+    gate_up: dict[int, torch.Tensor] = {}
+    hooks = []
+    modulated_forwards = {}
+    for index in wanted:
+        fc1 = model.backbone.blocks[index].mlp.fc1
+        hooks.append(fc1.register_forward_hook(
+            lambda _module, _inputs, output, index=index: gate_up.__setitem__(index, output.detach().clone())
+        ))
+        original = fc1.forward_modulated
+        modulated_forwards[index] = original
+
+        def capture_modulated(_self, *values, index=index, original=original, **kwargs):
+            output = original(*values, **kwargs)
+            gate_up[index] = output.detach().clone()
+            return output
+
+        fc1.forward_modulated = types.MethodType(capture_modulated, fc1)
+    try:
+        with torch.inference_mode():
+            for index, (block, adaln) in enumerate(zip(model.backbone.blocks, model.backbone.adaln, strict=True)):
+                if index in wanted:
+                    block_inputs[index] = hidden.detach().clone()
+                hidden = block(hidden, rotation, *adaln(timesteps), segments)
+                if index in wanted:
+                    block_outputs[index] = hidden.detach().clone()
+    finally:
+        for hook in hooks:
+            hook.remove()
+        for index, original in modulated_forwards.items():
+            model.backbone.blocks[index].mlp.fc1.forward_modulated = original
+    if set(gate_up) != wanted:
+        raise RuntimeError(f"failed to capture FC1 boundaries: got {sorted(gate_up)}")
+    return block_inputs, block_outputs, gate_up
+
+
+def packed_boundary(gate_up: torch.Tensor, fc2):
+    tensor_scale, qdata, block_scale = vortex_native_quantize_swiglu_nvfp4(gate_up)
+    alpha = (tensor_scale.float() * fc2.weight_scale_2.float()).reshape(1).contiguous()
+    beta = torch.zeros(1, device=gate_up.device, dtype=torch.float32)
+    return tensor_scale, qdata, block_scale, alpha, beta
+
+
+def baseline_metadata(fc2, gate_up: torch.Tensor) -> dict[str, Any]:
+    from torch.profiler import ProfilerActivity, profile
+
+    with torch.inference_mode():
+        fc2.forward_swiglu(gate_up)
+        torch.cuda.synchronize()
+        with profile(activities=[ProfilerActivity.CPU, ProfilerActivity.CUDA]) as captured:
+            output = fc2.forward_swiglu(gate_up)
+        torch.cuda.synchronize()
+    events = []
+    for event in captured.events():
+        if "CUDA" not in str(getattr(event, "device_type", "")):
+            continue
+        device_us = max(
+            float(getattr(event, "device_time_total", 0.0) or 0.0),
+            float(getattr(event, "self_device_time_total", 0.0) or 0.0),
+            float(getattr(event, "cuda_time_total", 0.0) or 0.0),
+            float(getattr(event, "self_cuda_time_total", 0.0) or 0.0),
+        )
+        events.append({"name": event.name, "device_time_total_us": device_us})
+    events.sort(key=lambda row: row["device_time_total_us"], reverse=True)
+    return {
+        "path": "fc2.forward_swiglu -> accepted producer -> Comfy Kitchen 0.2.31 scaled_mm_nvfp4",
+        "descriptors": {
+            "packed_input_output": "row-major [M,K] @ [N,K].T -> BF16 [M,N]",
+            "block_scale_mode": "VEC16_UE4M3",
+            "compute_and_scale": "FP32",
+            "scalar_pointer_mode": "device",
+            "bias": None,
+            "beta": 0.0,
+            "comfy_kitchen_version": "0.2.31",
+        },
+        "profiler_cuda_events_available": bool(events),
+        "profiler_note": None if events else "Torch profiler returned no CUDA kernel events on this build; use --mode profile with NCU for kernel metadata.",
+        "top_cuda_events": events[:10],
+        "output_sha256": tensor_sha256(output.detach().clone()),
+    }
+
+
+def enumerate_all(extension, packed, fc2, args: argparse.Namespace):
+    _tensor_scale, qdata, block_scale, _alpha, _beta = packed
+    groups = []
+    errors = []
+    for budget in args.workspace_budgets:
+        try:
+            values = [dict(value) for value in extension.enumerate(
+                qdata, block_scale, fc2.weight, fc2.weight_scale, budget, args.requested_count,
+            )]
+            groups.append({"max_workspace_bytes": budget, "requested_count": args.requested_count, "returned_count": len(values), "algorithms": values})
+        except Exception as error:
+            groups.append({"max_workspace_bytes": budget, "requested_count": args.requested_count, "returned_count": 0, "algorithms": []})
+            errors.append({"operation": "heuristic", "max_workspace_bytes": budget, "error": repr(error)})
+    return groups, errors
+
+
+def explicit_checks(extension, packed, fc2, enumerated, args: argparse.Namespace):
+    _tensor_scale, qdata, block_scale, _alpha, _beta = packed
+    unique_bases: dict[int, dict[str, Any]] = {}
+    for group in enumerated:
+        for candidate in group["algorithms"]:
+            unique_bases.setdefault(candidate["algorithm_id"], candidate)
+    checked, errors = [], []
+    attempts = 0
+    for base in unique_bases.values():
+        mask = int(base.get("capabilities", {}).get("reduction_scheme_mask", 0))
+        reductions = [1 << bit for bit in range(32) if mask & (1 << bit)]
+        for split_k in args.explicit_split_k:
+            schemes = [0] if split_k == 1 else reductions
+            if not schemes:
+                errors.append({"algorithm_id": base["algorithm_id"], "split_k": split_k, "error": "capability reports no public reduction scheme"})
+            for reduction in schemes:
+                if attempts >= args.max_explicit_checks:
+                    return checked, errors
+                attempts += 1
+                config = {key: base[key] for key in ("algorithm_id", "tile_id", "stages_id", "custom_option", "cta_swizzle", "inner_shape", "cluster_shape") if key in base}
+                config.update({"split_k": split_k, "reduction_scheme": reduction})
+                try:
+                    result = dict(extension.check(qdata, block_scale, fc2.weight, fc2.weight_scale, config))
+                    result["requested_config"] = config
+                    checked.append(result)
+                except Exception as error:
+                    errors.append({"config": config, "error": repr(error)})
+    return checked, errors
+
+
+def resolve_candidate(args: argparse.Namespace, enumerated, explicit) -> dict[str, Any]:
+    if args.candidate:
+        if args.candidate.startswith("enumerate:"):
+            _, group, result = args.candidate.split(":")
+            return dict(enumerated[int(group)]["algorithms"][int(result)])
+        path = Path(args.candidate)
+        return json.loads(path.read_text()) if path.exists() else json.loads(args.candidate)
+    valid_explicit = [row for row in explicit if row.get("valid") and row.get("required_workspace_bytes", sys.maxsize) <= args.workspace_bytes]
+    if valid_explicit:
+        return valid_explicit[0]["requested_config"]
+    for group in enumerated:
+        for row in group["algorithms"]:
+            if row.get("valid") and row.get("required_workspace_bytes", sys.maxsize) <= args.workspace_bytes:
+                return row
+    raise RuntimeError("no valid candidate fits --workspace-bytes")
+
+
+def candidate_runner(extension, packed, fc2, config, workspace_bytes: int):
+    _tensor_scale, qdata, block_scale, alpha, beta = packed
+    output = torch.empty((qdata.shape[0], fc2.out_features), device=qdata.device, dtype=torch.bfloat16)
+    workspace = torch.empty(workspace_bytes, device=qdata.device, dtype=torch.uint8)
+
+    def run() -> torch.Tensor:
+        extension.run(qdata, block_scale, fc2.weight, fc2.weight_scale, alpha, beta, output, workspace, config)
+        return output
+
+    return run, output, workspace
+
+
+def comfy_packed_runner(packed, fc2, logical_rows: int):
+    import torch.nn.functional as functional
+    from comfy_kitchen.tensor import QuantizedTensor, TensorCoreNVFP4Layout
+
+    tensor_scale, qdata, block_scale, _alpha, _beta = packed
+    activation = QuantizedTensor(
+        qdata,
+        "TensorCoreNVFP4Layout",
+        TensorCoreNVFP4Layout.Params(
+            scale=tensor_scale,
+            orig_dtype=torch.bfloat16,
+            orig_shape=(logical_rows, fc2.in_features),
+            block_scale=block_scale,
+        ),
+    )
+    weight = fc2._packed_weight()
+
+    def run():
+        return functional.linear(activation, weight, None)[:logical_rows, :fc2.out_features]
+
+    return run
+
+
+def benchmark_pair(baseline, candidate, rounds: int, warmup: int, m: int, n: int, k: int, *, report_dense: bool = True):
+    with torch.inference_mode():
+        for _ in range(warmup):
+            baseline()
+            candidate()
+        torch.cuda.synchronize()
+        samples = {"baseline": [], "candidate": []}
+        last = {}
+        for round_index in range(rounds):
+            order = (("baseline", baseline), ("candidate", candidate)) if round_index % 2 == 0 else (("candidate", candidate), ("baseline", baseline))
+            for name, call in order:
+                elapsed, output = timed_cuda(call)
+                samples[name].append(elapsed)
+                last[name] = output.detach().clone()
+    return {
+        "order": "AB/BA alternates by round",
+        "baseline": timing_summary(samples["baseline"], m if report_dense else None, n if report_dense else None, k if report_dense else None),
+        "candidate": timing_summary(samples["candidate"], m if report_dense else None, n if report_dense else None, k if report_dense else None),
+        "parity": compare(last["candidate"][:m, :n], last["baseline"][:m, :n]),
+    }
+
+
+def benchmark_fc2_candidate(extension, gate_up, fc2, packed, config, args):
+    run_candidate, output, workspace = candidate_runner(extension, packed, fc2, config, args.workspace_bytes)
+    m, n, k = gate_up.shape[0], fc2.out_features, fc2.in_features
+    run_comfy_packed = comfy_packed_runner(packed, fc2, m)
+    direct = benchmark_pair(
+        run_comfy_packed, run_candidate,
+        args.rounds, args.warmup, m, n, k,
+    )
+
+    boundary_output = torch.empty_like(output)
+    boundary_workspace = torch.empty_like(workspace)
+
+    def candidate_boundary():
+        fresh = packed_boundary(gate_up, fc2)
+        extension.run(
+            fresh[1], fresh[2], fc2.weight, fc2.weight_scale, fresh[3], fresh[4],
+            boundary_output, boundary_workspace, config,
+        )
+        return boundary_output
+
+    boundary = benchmark_pair(
+        lambda: fc2.forward_swiglu(gate_up), candidate_boundary,
+        args.rounds, args.warmup, m, n, k,
+    )
+    if not direct["parity"]["bf16_exact"] or not boundary["parity"]["bf16_exact"]:
+        raise RuntimeError("FC2 library candidate is not byte-exact")
+    checked = dict(extension.check(packed[1], packed[2], fc2.weight, fc2.weight_scale, config))
+    return {
+        "selected_config": config,
+        "supplied_workspace_bytes": workspace.numel(),
+        "required_workspace_bytes": checked.get("required_workspace_bytes"),
+        "checked": checked,
+        "fc2_only": direct,
+        "accepted_producer_plus_fc2": boundary,
+        "output_sha256": tensor_sha256(output[:m].detach().clone()),
+    }
+
+
+def profile_one(extension, packed, fc2, config, logical_rows: int, args):
+    run_candidate, output, workspace = candidate_runner(extension, packed, fc2, config, args.workspace_bytes)
+    expected = None
+    if args.profile_target == "baseline":
+        run_profiled = comfy_packed_runner(packed, fc2, logical_rows)
+    else:
+        run_profiled = run_candidate
+    with torch.inference_mode():
+        for _ in range(args.warmup):
+            run_profiled()
+        if args.profile_target == "candidate":
+            expected = comfy_packed_runner(packed, fc2, logical_rows)().detach().clone()
+        torch.cuda.synchronize()
+        torch.cuda.cudart().cudaProfilerStart()
+        output = run_profiled()
+        torch.cuda.cudart().cudaProfilerStop()
+        torch.cuda.synchronize()
+    parity = compare(output[:logical_rows], expected[:logical_rows]) if expected is not None else None
+    if parity is not None and not parity["bf16_exact"]:
+        raise RuntimeError("profiled FC2 library candidate is not byte-exact")
+    return {
+        "target": args.profile_target,
+        "selected_config": config if args.profile_target == "candidate" else None,
+        "supplied_workspace_bytes": workspace.numel() if args.profile_target == "candidate" else 32 << 20,
+        "output_sha256": tensor_sha256(output[:logical_rows, :fc2.out_features]),
+        "candidate_vs_baseline": parity,
+    }
+
+
+def block_gate(extension, model, block_inputs, block_outputs, gate_up, timesteps, rotation, segments, config, args):
+    rows = []
+    for index in DEFAULT_BLOCKS:
+        block = model.backbone.blocks[index]
+        fc2 = block.mlp.fc2
+        packed = packed_boundary(gate_up[index], fc2)
+        _tensor_scale, qdata, _block_scale, _alpha, beta = packed
+        candidate_output = torch.empty(
+            (qdata.shape[0], fc2.out_features), device=qdata.device, dtype=torch.bfloat16,
+        )
+        workspace = torch.empty(args.workspace_bytes, device=qdata.device, dtype=torch.uint8)
+        original = fc2.forward_swiglu
+
+        def replacement(_self, actual_gate_up, expected=gate_up[index]):
+            if actual_gate_up.shape != expected.shape:
+                raise RuntimeError("block-gate FC2 received an unexpected boundary shape")
+            tensor_scale, actual_qdata, actual_block_scale = vortex_native_quantize_swiglu_nvfp4(actual_gate_up)
+            alpha = (tensor_scale.float() * fc2.weight_scale_2.float()).reshape(1).contiguous()
+            extension.run(
+                actual_qdata, actual_block_scale, fc2.weight, fc2.weight_scale,
+                alpha, beta, candidate_output, workspace, config,
+            )
+            return candidate_output[:actual_gate_up.shape[0], :fc2.out_features]
+
+        candidate_method = types.MethodType(replacement, fc2)
+        adaln_values = tuple(value.detach().clone() for value in model.backbone.adaln[index](timesteps))
+
+        def baseline():
+            fc2.forward_swiglu = original
+            return block(block_inputs[index].detach().clone(), rotation, *adaln_values, segments)
+
+        def candidate():
+            fc2.forward_swiglu = candidate_method
+            return block(block_inputs[index].detach().clone(), rotation, *adaln_values, segments)
+
+        try:
+            timing = benchmark_pair(
+                baseline, candidate, args.rounds, args.warmup,
+                block_outputs[index].shape[0], block_outputs[index].shape[1], 1,
+                report_dense=False,
+            )
+            baseline_value = baseline().detach().clone()
+            candidate_value = candidate().detach().clone()
+        finally:
+            fc2.forward_swiglu = original
+        row = {
+            "block": index,
+            "only_monkeypatched_method": "block.mlp.fc2.forward_swiglu",
+            "accepted_gate_and_residual_path_preserved": True,
+            "candidate_vs_baseline": compare(candidate_value, baseline_value),
+            "baseline_vs_traversal": compare(baseline_value, block_outputs[index]),
+            "candidate_vs_traversal": compare(candidate_value, block_outputs[index]),
+            "timing": timing,
+            "supplied_workspace_bytes": workspace.numel(),
+        }
+        if not all(
+            row[name]["bf16_exact"]
+            for name in ("candidate_vs_baseline", "baseline_vs_traversal", "candidate_vs_traversal")
+        ):
+            raise RuntimeError(f"FC2 library candidate is not byte-exact in block {index}")
+        rows.append(row)
+    return rows
+
+
+def trajectory_gate(extension, checkpoint, model, config, args):
+    packer = H3PromptPacker(checkpoint)
+    torch.manual_seed(args.seed)
+    video, audio, aligned_frames = random_av_latents(
+        args.width, args.height, args.frames, args.seed, device="cuda",
+    )
+    text = torch.randn(
+        1, args.text_tokens, 5376, device="cuda", dtype=torch.bfloat16,
+    )
+    originals = [block.mlp.fc2.forward_swiglu for block in model.backbone.blocks]
+    shared_output = torch.empty(
+        (((37_810 + 15) // 16) * 16, 5376), device="cuda", dtype=torch.bfloat16,
+    )
+    workspace = torch.empty(args.workspace_bytes, device="cuda", dtype=torch.uint8)
+    beta = torch.zeros(1, device="cuda", dtype=torch.float32)
+
+    def candidate_method(fc2):
+        def replacement(_self, actual_gate_up):
+            tensor_scale, qdata, block_scale = vortex_native_quantize_swiglu_nvfp4(actual_gate_up)
+            if qdata.shape[0] > shared_output.shape[0] or fc2.out_features > shared_output.shape[1]:
+                raise RuntimeError("trajectory FC2 boundary exceeds the preallocated canonical output")
+            alpha = (tensor_scale.float() * fc2.weight_scale_2.float()).reshape(1).contiguous()
+            output = shared_output[:qdata.shape[0], :fc2.out_features]
+            extension.run(
+                qdata, block_scale, fc2.weight, fc2.weight_scale,
+                alpha, beta, output, workspace, config,
+            )
+            return output[:actual_gate_up.shape[0], :fc2.out_features]
+        return types.MethodType(replacement, fc2)
+
+    candidates = [candidate_method(block.mlp.fc2) for block in model.backbone.blocks]
+
+    def run(candidate: bool):
+        for index, block in enumerate(model.backbone.blocks):
+            block.mlp.fc2.forward_swiglu = candidates[index] if candidate else originals[index]
+        torch.cuda.synchronize()
+        started = time.perf_counter()
+        result = sample_video_res_multistep(
+            model,
+            packer,
+            text,
+            video.detach().clone(),
+            audio.detach().clone(),
+            steps=args.steps,
+            seed=args.seed,
+            return_audio=True,
+            progress=True,
+        )
+        torch.cuda.synchronize()
+        return result, time.perf_counter() - started
+
+    try:
+        with torch.inference_mode():
+            (reference_video, reference_audio), baseline_seconds = run(False)
+            (candidate_video, candidate_audio), candidate_seconds = run(True)
+    finally:
+        for block, original in zip(model.backbone.blocks, originals, strict=True):
+            block.mlp.fc2.forward_swiglu = original
+
+    video_parity = compare(candidate_video, reference_video)
+    audio_parity = compare(candidate_audio, reference_audio)
+    result = {
+        "steps": args.steps,
+        "resolution": [args.width, args.height],
+        "frames": aligned_frames,
+        "seed": args.seed,
+        "baseline_seconds": baseline_seconds,
+        "candidate_seconds": candidate_seconds,
+        "improvement_percent": (1.0 - candidate_seconds / baseline_seconds) * 100.0,
+        "video_parity": video_parity,
+        "audio_parity": audio_parity,
+        "bf16_exact": video_parity["bf16_exact"] and audio_parity["bf16_exact"],
+        "selected_config": config,
+        "supplied_workspace_bytes": workspace.numel(),
+        "all_50_fc2_calls_replaced": True,
+        "accepted_swiglu_producer_preserved": True,
+        "accepted_gate_and_residual_path_preserved": True,
+    }
+    if not result["bf16_exact"]:
+        raise RuntimeError("FC2 library candidate trajectory is not byte-exact")
+    return result
+
+
+def main() -> None:
+    args = parse_args()
+    if not torch.cuda.is_available():
+        raise RuntimeError("CUDA is required")
+    extension = load_extension(args)
+    if args.mode == "compile":
+        result = {"mode": "compile", "extension": dict(extension.build_info())}
+        print(json.dumps(result, indent=2), flush=True)
+        return
+
+    checkpoint = H3Checkpoint(args.model_path, device="cuda")
+    model = H3PackedDenoiser.from_checkpoint(checkpoint, attention_backend="sage2").eval()
+    hidden, timesteps, rotation, segments, workload = make_workload(args, checkpoint, model)
+    with torch.inference_mode():
+        block_inputs, block_outputs, gate_ups = capture_boundaries(
+            args, model, hidden, timesteps, rotation, segments,
+        )
+    probe = args.probe_block
+    if probe not in gate_ups:
+        raise ValueError("--probe-block must be one of the retained blocks")
+    fc2 = model.backbone.blocks[probe].mlp.fc2
+    if fc2.bias is not None:
+        raise RuntimeError("this no-bias FC2 scheduling study refuses a biased module")
+    packed = packed_boundary(gate_ups[probe], fc2)
+    enumerated, errors = enumerate_all(extension, packed, fc2, args)
+    explicit, explicit_errors = explicit_checks(extension, packed, fc2, enumerated, args)
+    errors.extend(explicit_errors)
+    result: dict[str, Any] = {
+        "mode": args.mode,
+        "environment": environment(args, extension),
+        "workload": workload,
+        "retained_blocks": list(DEFAULT_BLOCKS),
+        "immutable_cloned_block_inputs": {str(index): list(value.shape) for index, value in block_inputs.items()},
+        "fc2_boundary": {
+            "block": probe,
+            "gate_up_shape": list(gate_ups[probe].shape),
+            "activation_qdata_shape": list(packed[1].shape),
+            "weight_qdata_shape": list(fc2.weight.shape),
+            "logical_mnk": [gate_ups[probe].shape[0], fc2.out_features, fc2.in_features],
+            "descriptor_mnk_after_padding": [packed[1].shape[0], fc2.out_features, fc2.in_features],
+            "producer": "vortex_native_quantize_swiglu_nvfp4",
+            "no_bias": fc2.bias is None,
+        },
+        "baseline_kernel_metadata": baseline_metadata(fc2, gate_ups[probe]),
+        "heuristics": enumerated,
+        "explicit_split_k_checks": explicit,
+    }
+
+    if args.mode in {"characterize", "sweep", "selected", "profile", "block-gate", "trajectory"}:
+        config = resolve_candidate(args, enumerated, explicit)
+        result["selected"] = config
+    if args.mode == "characterize":
+        result["candidate_probe"] = benchmark_fc2_candidate(extension, gate_ups[probe], fc2, packed, config, args)
+    elif args.mode == "sweep":
+        candidates = []
+        seen = set()
+        pool = [row for group in enumerated for row in group["algorithms"]] + [row for row in explicit if row.get("valid")]
+        for row in pool:
+            candidate = row.get("requested_config", row)
+            key = tuple(candidate.get(name) for name in ("algorithm_id", "tile_id", "stages_id", "split_k", "reduction_scheme", "custom_option", "cta_swizzle", "inner_shape", "cluster_shape"))
+            if key in seen or row.get("required_workspace_bytes", 0) > args.workspace_bytes:
+                continue
+            seen.add(key)
+            try:
+                candidates.append(benchmark_fc2_candidate(extension, gate_ups[probe], fc2, packed, candidate, args))
+            except Exception as error:
+                errors.append({"config": candidate, "operation": "benchmark", "error": repr(error)})
+            if len(candidates) >= args.candidate_limit:
+                break
+        result["candidates"] = candidates
+    elif args.mode == "selected":
+        result["candidate_probe"] = benchmark_fc2_candidate(extension, gate_ups[probe], fc2, packed, config, args)
+    elif args.mode == "profile":
+        result["profile"] = profile_one(extension, packed, fc2, config, gate_ups[probe].shape[0], args)
+    elif args.mode == "block-gate":
+        result["block_gate"] = block_gate(extension, model, block_inputs, block_outputs, gate_ups, timesteps, rotation, segments, config, args)
+    elif args.mode == "trajectory":
+        result["trajectory"] = trajectory_gate(extension, checkpoint, model, config, args)
+
+    result["errors_and_unsupported"] = errors + [{
+        "feature": "Stream-K",
+        "supported_public_control": False,
+        "reason": extension.build_info()["stream_k_note"],
+    }]
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    args.output.write_text(json.dumps(result, indent=2) + "\n", encoding="utf-8")
+    print(json.dumps(result, indent=2), flush=True)
+
+
+if __name__ == "__main__":
+    main()