{ "device": "NVIDIA GB10", "cutlass_dsl": "4.6.2", "metadata": { "width": 1344, "height": 768, "frames": 124, "steps": 12, "sampler_step": 1, "seed": 440420, "text_tokens": 100, "hidden_shape": [ 37810, 5376 ], "segments": [ [ 0, 100, 1 ], [ 100, 514, 2 ], [ 514, 37810, 0 ] ] }, "linear": "mlp_fc1", "mnk": [ 128, 28672, 5376 ], "ring": { "row_capacity": 128, "producer": "vortex_native_quantize_nvfp4", "qdata_bytes": 344064, "sfa_bytes": 43008, "logical_bytes": 387072 }, "parity": { "tensor_scale_equal": true, "fp4_difference_count": 0, "block_scale_difference_count": 0, "output_equal": true, "max_abs": 0.0, "mean_abs": 0.0 }, "timing": { "warmup": 10, "iterations": 100, "producer_ms": 0.015716160535812377, "global_scale_ms": 0.014631999731063843, "cute_consumer_ms": 0.6108745574951172, "modeled_ring_ms": 0.6412227177619935, "actual_into_ring_cute_gemm_ms": 0.6573414611816406, "reference_vortex_scale_comfy_pack_gemm_ms": 0.5960015869140625, "modeled_ring_vs_reference": 1.0758741785941779, "actual_ring_vs_reference": 1.1029189780939672 } }