{ "prompt": "A playful orange tabby cat starts in an ordinary cozy living room in a normal house, afternoon light, sofa and rug. The cat crouches, jumps, and does one clean athletic backflip in slow motion. As the backflip completes there is a sharp cinematic cut: the cat lands perfectly on a glowing neon disco dance floor wearing oversized black sunglasses. Mirror ball reflections, colorful lights, joyful party energy, stylish and funny, clear before-and-after transformation.", "model_path": "/models/minimax_h3_fl2va_pruned_nvfp4.safetensors", "width": 1344, "height": 768, "frames": 124, "steps": 12, "sampler_step": 1, "seed": 440420, "text_tokens": 100, "block_index": 24, "attention": "sage2", "warmup": 5, "iterations": 10, "hidden_shape": [ 37810, 5376 ], "output_shape": [ 37810, 5376 ], "segments": [ [ 0, 100, 1 ], [ 100, 514, 2 ], [ 514, 37810, 0 ] ], "timings": { "norm1": { "count": 10, "mean_s": 0.0036379225028213115, "p50_s": 0.003614433022448793, "p90_s": 0.0037366025615483524, "p95_s": 0.0037547897722106423, "p99_s": 0.003769339540740475, "min_s": 0.0035753760021179914, "max_s": 0.003772976982872933 }, "modulate_msa": { "count": 10, "mean_s": 0.010608601296553388, "p50_s": 0.010585986980004236, "p90_s": 0.011015687783947214, "p95_s": 0.011020101406029426, "p99_s": 0.011023632303695194, "min_s": 0.01028603402664885, "max_s": 0.011024515028111637 }, "linear.attn_qkv_proj.flatten_contiguous": { "count": 10, "mean_s": 4.382384940981865e-06, "p50_s": 4.103989340364933e-06, "p90_s": 4.502356750890611e-06, "p95_s": 5.7551631471142146e-06, "p99_s": 6.757408264093101e-06, "min_s": 3.95200913771987e-06, "max_s": 7.007969543337822e-06 }, "linear.attn_qkv_proj.pre_quant_scale": { "count": 10, "mean_s": 0.0, "p50_s": 0.0, "p90_s": 0.0, "p95_s": 0.0, "p99_s": 0.0, "min_s": 0.0, "max_s": 0.0 }, "linear.attn_qkv_proj.packed_weight_wrapper": { "count": 10, "mean_s": 1.015679445117712e-05, "p50_s": 8.024013368412852e-06, "p90_s": 1.037599868141114e-05, "p95_s": 1.9915992743335642e-05, "p99_s": 2.754798799287528e-05, "min_s": 7.760012522339821e-06, "max_s": 2.945598680526018e-05 }, "linear.attn_qkv_proj.bias_cast": { "count": 10, "mean_s": 3.4719996619969606e-06, "p50_s": 3.416003892198205e-06, "p90_s": 3.587204264476895e-06, "p95_s": 3.7455960409715767e-06, "p99_s": 3.872309462167323e-06, "min_s": 3.328023012727499e-06, "max_s": 3.903987817466259e-06 }, "linear.attn_qkv_proj.activation_scale": { "count": 10, "mean_s": 0.001784054801100865, "p50_s": 0.0017840564833022654, "p90_s": 0.0018375497369561345, "p95_s": 0.001851214884663932, "p99_s": 0.0018621470028301699, "min_s": 0.0017313450225628912, "max_s": 0.0018648800323717296 }, "linear.attn_qkv_proj.scale_to_device": { "count": 10, "mean_s": 1.0225595906376838e-05, "p50_s": 9.207986295223236e-06, "p90_s": 1.1627195635810492e-05, "p95_s": 1.5061601880006484e-05, "p99_s": 1.7809126875363292e-05, "min_s": 8.128001354634762e-06, "max_s": 1.849600812420249e-05 }, "linear.attn_qkv_proj.activation_quant_pack": { "count": 10, "mean_s": 0.002209685498382896, "p50_s": 0.0021846969902981073, "p90_s": 0.0022998010332230478, "p95_s": 0.0023025010275887327, "p99_s": 0.0023046610230812803, "min_s": 0.0021451999782584608, "max_s": 0.0023052010219544172 }, "linear.attn_qkv_proj.activation_quant_wrap": { "count": 10, "mean_s": 6.39361678622663e-06, "p50_s": 6.1680038925260305e-06, "p90_s": 6.588839460164307e-06, "p95_s": 7.510429713875053e-06, "p99_s": 8.247701916843653e-06, "min_s": 5.936017259955406e-06, "max_s": 8.432019967585802e-06 }, "linear.attn_qkv_proj.gemm": { "count": 10, "mean_s": 0.026259651698637754, "p50_s": 0.026372870983323082, "p90_s": 0.026779549376806246, "p95_s": 0.026793114203610458, "p99_s": 0.02680396606505383, "min_s": 0.025480758980847895, "max_s": 0.02680667903041467 }, "linear.attn_qkv_proj.slice_reshape": { "count": 10, "mean_s": 5.59519394300878e-06, "p50_s": 5.456007784232497e-06, "p90_s": 5.7215744163841006e-06, "p95_s": 6.412793300114571e-06, "p99_s": 6.965768407098949e-06, "min_s": 5.264009814709425e-06, "max_s": 7.104012183845043e-06 }, "attn_qkv_proj": { "count": 10, "mean_s": 0.03033423840533942, "p50_s": 0.03052276000380516, "p90_s": 0.030788672040216625, "p95_s": 0.03079540401813574, "p99_s": 0.030800789600471036, "min_s": 0.02970914397155866, "max_s": 0.030802135996054858 }, "attn_qkv_split_view": { "count": 10, "mean_s": 1.1774385347962379e-05, "p50_s": 1.0199990356341004e-05, "p90_s": 1.3715191744267938e-05, "p95_s": 1.768957590684294e-05, "p99_s": 2.0869083236902955e-05, "min_s": 9.935989510267973e-06, "max_s": 2.1663960069417953e-05 }, "attn_qk_rms_rope": { "count": 10, "mean_s": 0.012174534396035597, "p50_s": 0.012181266996776685, "p90_s": 0.012462468614103273, "p95_s": 0.012474859817302786, "p99_s": 0.012484772779862397, "min_s": 0.011805346992332488, "max_s": 0.012487251020502299 }, "attention_kernel": { "count": 10, "mean_s": 0.26265805189614183, "p50_s": 0.26240472550853156, "p90_s": 0.2634901970799547, "p95_s": 0.26394581352651586, "p99_s": 0.2643103066837648, "min_s": 0.2618118610116653, "max_s": 0.26440142997307703 }, "attn_output_reshape": { "count": 10, "mean_s": 4.099193029105664e-06, "p50_s": 4.024012014269829e-06, "p90_s": 4.249595804139972e-06, "p95_s": 4.508779966272413e-06, "p99_s": 4.716127295978368e-06, "min_s": 3.840017598122358e-06, "max_s": 4.767964128404856e-06 }, "linear.attn_out_proj.flatten_contiguous": { "count": 10, "mean_s": 4.086404806002974e-06, "p50_s": 4.064000677317381e-06, "p90_s": 4.27677296102047e-06, "p95_s": 4.370388342067599e-06, "p99_s": 4.445280646905303e-06, "min_s": 3.9199949242174625e-06, "max_s": 4.464003723114729e-06 }, "linear.attn_out_proj.pre_quant_scale": { "count": 10, "mean_s": 0.0, "p50_s": 0.0, "p90_s": 0.0, "p95_s": 0.0, "p99_s": 0.0, "min_s": 0.0, "max_s": 0.0 }, "linear.attn_out_proj.packed_weight_wrapper": { "count": 10, "mean_s": 8.564791642129421e-06, "p50_s": 8.360017091035843e-06, "p90_s": 8.947186870500445e-06, "p95_s": 9.249590220861137e-06, "p99_s": 9.49151290114969e-06, "min_s": 8.192029781639576e-06, "max_s": 9.551993571221828e-06 }, "linear.attn_out_proj.bias_cast": { "count": 10, "mean_s": 3.399985143914819e-06, "p50_s": 3.3919932320713997e-06, "p90_s": 3.489578375592828e-06, "p95_s": 3.4967815736308694e-06, "p99_s": 3.5025441320613027e-06, "min_s": 3.263994585722685e-06, "max_s": 3.503984771668911e-06 }, "linear.attn_out_proj.activation_scale": { "count": 10, "mean_s": 0.002312088292092085, "p50_s": 0.0023111284826882184, "p90_s": 0.0023206335958093406, "p95_s": 0.0023220447939820586, "p99_s": 0.002323173752520233, "min_s": 0.0022997119813226163, "max_s": 0.002323455992154777 }, "linear.attn_out_proj.scale_to_device": { "count": 10, "mean_s": 9.87689709290862e-06, "p50_s": 9.71200643107295e-06, "p90_s": 1.0727980406954884e-05, "p95_s": 1.0835975990630686e-05, "p99_s": 1.0922372457571327e-05, "min_s": 9.34401759877801e-06, "max_s": 1.0943971574306488e-05 }, "linear.attn_out_proj.activation_quant_pack": { "count": 10, "mean_s": 0.0028904856008011848, "p50_s": 0.0028887049993500113, "p90_s": 0.00291029930813238, "p95_s": 0.002919062168803066, "p99_s": 0.0029260724573396146, "min_s": 0.002867089002393186, "max_s": 0.002927825029473752 }, "linear.attn_out_proj.activation_quant_wrap": { "count": 10, "mean_s": 6.348802708089352e-06, "p50_s": 6.263988325372338e-06, "p90_s": 6.524822674691677e-06, "p95_s": 6.726407445967197e-06, "p99_s": 6.887675262987614e-06, "min_s": 5.999987479299307e-06, "max_s": 6.927992217242718e-06 }, "linear.attn_out_proj.gemm": { "count": 10, "mean_s": 0.007921349199023098, "p50_s": 0.00793309000437148, "p90_s": 0.007959806796861812, "p95_s": 0.007965408396557905, "p99_s": 0.007969889676314778, "min_s": 0.007834289979655296, "max_s": 0.007971009996253997 }, "linear.attn_out_proj.slice_reshape": { "count": 10, "mean_s": 5.4496107622981075e-06, "p50_s": 5.368026904761791e-06, "p90_s": 5.625589983537793e-06, "p95_s": 5.740814958699047e-06, "p99_s": 5.832994938828052e-06, "min_s": 5.311972927302122e-06, "max_s": 5.856039933860302e-06 }, "attn_out_proj": { "count": 10, "mean_s": 0.01320089278742671, "p50_s": 0.013211563491495326, "p90_s": 0.013256034994265066, "p95_s": 0.01325833899609279, "p99_s": 0.01326018219755497, "min_s": 0.013092691951896995, "max_s": 0.013260642997920513 }, "gate_msa": { "count": 10, "mean_s": 0.008729595987824723, "p50_s": 0.00866641046013683, "p90_s": 0.008878554130205885, "p95_s": 0.008967942543677054, "p99_s": 0.00903945327445399, "min_s": 0.008640034007839859, "max_s": 0.009057330957148224 }, "norm2": { "count": 10, "mean_s": 0.003563264914555475, "p50_s": 0.0035559765237849206, "p90_s": 0.0035853242268785836, "p95_s": 0.0036025826178956773, "p99_s": 0.0036163893307093534, "min_s": 0.003548705019056797, "max_s": 0.003619841008912772 }, "modulate_mlp": { "count": 10, "mean_s": 0.010596700402675196, "p50_s": 0.010504154517548159, "p90_s": 0.011042852600803598, "p95_s": 0.01105070777994115, "p99_s": 0.01105699192325119, "min_s": 0.01029996998840943, "max_s": 0.0110585629590787 }, "linear.mlp_fc1.flatten_contiguous": { "count": 10, "mean_s": 4.132796311751008e-06, "p50_s": 4.064029781147838e-06, "p90_s": 4.33918321505189e-06, "p95_s": 4.5695807784795754e-06, "p99_s": 4.7538988292217255e-06, "min_s": 3.936002030968666e-06, "max_s": 4.799978341907263e-06 }, "linear.mlp_fc1.pre_quant_scale": { "count": 10, "mean_s": 0.0, "p50_s": 0.0, "p90_s": 0.0, "p95_s": 0.0, "p99_s": 0.0, "min_s": 0.0, "max_s": 0.0 }, "linear.mlp_fc1.packed_weight_wrapper": { "count": 10, "mean_s": 7.998401997610927e-06, "p50_s": 7.86400050856173e-06, "p90_s": 8.32958030514419e-06, "p95_s": 8.876787615008652e-06, "p99_s": 9.314553462900221e-06, "min_s": 7.3919654823839664e-06, "max_s": 9.423994924873114e-06 }, "linear.mlp_fc1.bias_cast": { "count": 10, "mean_s": 3.448012284934521e-06, "p50_s": 3.4080003388226032e-06, "p90_s": 3.6080251447856423e-06, "p95_s": 3.6440149415284394e-06, "p99_s": 3.6728067789226773e-06, "min_s": 3.360037226229906e-06, "max_s": 3.6800047382712364e-06 }, "linear.mlp_fc1.activation_scale": { "count": 10, "mean_s": 0.0017999590025283396, "p50_s": 0.001780224498361349, "p90_s": 0.0018680152890738098, "p95_s": 0.0018741996405879037, "p99_s": 0.0018791471217991784, "min_s": 0.0017290239920839667, "max_s": 0.0018803839921019971 }, "linear.mlp_fc1.scale_to_device": { "count": 10, "mean_s": 1.1595204705372453e-05, "p50_s": 9.039998985826969e-06, "p90_s": 1.3588822912424795e-05, "p95_s": 2.3762410273775437e-05, "p99_s": 3.1901280162855987e-05, "min_s": 8.27200710773468e-06, "max_s": 3.3935997635126114e-05 }, "linear.mlp_fc1.activation_quant_pack": { "count": 10, "mean_s": 0.002238086803117767, "p50_s": 0.002225264994194731, "p90_s": 0.0023277984117157755, "p95_s": 0.0023365392175037415, "p99_s": 0.002343531862134114, "min_s": 0.002152128960005939, "max_s": 0.002345280023291707 }, "linear.mlp_fc1.activation_quant_wrap": { "count": 10, "mean_s": 6.2112114392220976e-06, "p50_s": 6.256013875827193e-06, "p90_s": 6.483227480202913e-06, "p95_s": 6.4976338762789965e-06, "p99_s": 6.509158993139863e-06, "min_s": 5.839974619448185e-06, "max_s": 6.51204027235508e-06 }, "linear.mlp_fc1.gemm": { "count": 10, "mean_s": 0.03520601270138286, "p50_s": 0.0351380895008333, "p90_s": 0.03551860520965419, "p95_s": 0.035628779590479095, "p99_s": 0.03571691909513902, "min_s": 0.03475844202330336, "max_s": 0.035738953971304 }, "linear.mlp_fc1.slice_reshape": { "count": 10, "mean_s": 5.545601015910506e-06, "p50_s": 5.504000000655651e-06, "p90_s": 5.809619324281812e-06, "p95_s": 5.816822522319853e-06, "p99_s": 5.822585080750287e-06, "min_s": 5.2480027079582214e-06, "max_s": 5.824025720357895e-06 }, "mlp_fc1": { "count": 10, "mean_s": 0.03932547768927179, "p50_s": 0.03926424199016765, "p90_s": 0.03974540139315649, "p95_s": 0.03980256220092997, "p99_s": 0.03984829084714874, "min_s": 0.03897794696968049, "max_s": 0.03985972300870344 }, "mlp_swiglu": { "count": 10, "mean_s": 0.02429507830529474, "p50_s": 0.024255126510979608, "p90_s": 0.024437257228419183, "p95_s": 0.024547863617772236, "p99_s": 0.02463634872925468, "min_s": 0.024114214000292122, "max_s": 0.024658470007125288 }, "linear.mlp_fc2.flatten_contiguous": { "count": 10, "mean_s": 4.006398376077413e-06, "p50_s": 3.951980033889413e-06, "p90_s": 4.184030694887042e-06, "p95_s": 4.364005872048437e-06, "p99_s": 4.507986013777554e-06, "min_s": 3.824010491371155e-06, "max_s": 4.543981049209833e-06 }, "linear.mlp_fc2.pre_quant_scale": { "count": 10, "mean_s": 0.0, "p50_s": 0.0, "p90_s": 0.0, "p95_s": 0.0, "p99_s": 0.0, "min_s": 0.0, "max_s": 0.0 }, "linear.mlp_fc2.packed_weight_wrapper": { "count": 10, "mean_s": 7.497583283111453e-06, "p50_s": 7.327966159209609e-06, "p90_s": 8.059164974838495e-06, "p95_s": 8.18159314803779e-06, "p99_s": 8.279535686597228e-06, "min_s": 7.167982403188944e-06, "max_s": 8.304021321237087e-06 }, "linear.mlp_fc2.bias_cast": { "count": 10, "mean_s": 3.446405753493309e-06, "p50_s": 3.4240074455738068e-06, "p90_s": 3.515236312523484e-06, "p95_s": 3.565606311894953e-06, "p99_s": 3.6059023113921285e-06, "min_s": 3.3599790185689926e-06, "max_s": 3.6159763112664223e-06 }, "linear.mlp_fc2.activation_scale": { "count": 10, "mean_s": 0.004662481398554519, "p50_s": 0.0046446814958471805, "p90_s": 0.004776477930136025, "p95_s": 0.004784527962328866, "p99_s": 0.004790967988083139, "min_s": 0.004539761983323842, "max_s": 0.004792577994521707 }, "linear.mlp_fc2.scale_to_device": { "count": 10, "mean_s": 1.0712019866332412e-05, "p50_s": 9.048002539202571e-06, "p90_s": 1.111680176109075e-05, "p95_s": 1.866240054368971e-05, "p99_s": 2.4698879569768907e-05, "min_s": 8.144008461385965e-06, "max_s": 2.62079993262887e-05 }, "linear.mlp_fc2.activation_quant_pack": { "count": 10, "mean_s": 0.005886719806585461, "p50_s": 0.005925089004449546, "p90_s": 0.005984025914222002, "p95_s": 0.0060072454623877995, "p99_s": 0.0060258211009204385, "min_s": 0.005692336999345571, "max_s": 0.006030465010553598 }, "linear.mlp_fc2.activation_quant_wrap": { "count": 10, "mean_s": 6.070389645174146e-06, "p50_s": 6.079993909224868e-06, "p90_s": 6.268831202760339e-06, "p95_s": 6.398410187102855e-06, "p99_s": 6.502073374576867e-06, "min_s": 5.743990186601877e-06, "max_s": 6.52798917144537e-06 }, "linear.mlp_fc2.gemm": { "count": 10, "mean_s": 0.05435659209615551, "p50_s": 0.05434621399035677, "p90_s": 0.05462489890633151, "p95_s": 0.05483235294814222, "p99_s": 0.05499831618159078, "min_s": 0.053905503009445965, "max_s": 0.05503980698995292 }, "linear.mlp_fc2.slice_reshape": { "count": 10, "mean_s": 5.41119952686131e-06, "p50_s": 5.4480042308568954e-06, "p90_s": 5.521601997315884e-06, "p95_s": 5.60081098228693e-06, "p99_s": 5.664178170263767e-06, "min_s": 5.199981387704611e-06, "max_s": 5.6800199672579765e-06 }, "mlp_fc2": { "count": 10, "mean_s": 0.06498188290861436, "p50_s": 0.06497353702434339, "p90_s": 0.06526047951192596, "p95_s": 0.06543104073789437, "p99_s": 0.0655674897186691, "min_s": 0.06431148899719119, "max_s": 0.06560160196386278 }, "gate_mlp": { "count": 10, "mean_s": 0.0101405818015337, "p50_s": 0.008795610017841682, "p90_s": 0.010404527210630472, "p95_s": 0.016100218589417625, "p99_s": 0.020656771692447368, "min_s": 0.008680481987539679, "max_s": 0.021795909968204796 }, "block_total": { "count": 10, "mean_s": 0.4943276042060461, "p50_s": 0.492947026505135, "p90_s": 0.4960365816077683, "p95_s": 0.5006749098101864, "p99_s": 0.5043855723721208, "min_s": 0.4925488180015236, "max_s": 0.5053132380126044 } }, "module_forward": { "count": 10, "mean_s": 0.45834548950660975, "p50_s": 0.45877519852365367, "p90_s": 0.45995565400226046, "p95_s": 0.46014774998766367, "p99_s": 0.46030142677598634, "min_s": 0.4553232230246067, "max_s": 0.460339845973067 }, "module_forward_checksum": -895169280.0, "fused_elementwise": true, "profiler_summary": { "profiled_iterations": 2, "runtime_kernel_launches": 0, "runtime_kernel_launches_per_block": 0.0, "positive_self_device_allocated_bytes": 16019156992 }, "profiler_top_events": [ { "key": "aten::rms_norm", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1341.0399999999995, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], [], [5376], []]" }, { "key": "aten::_fused_rms_norm", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1330.4159999999995, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], [], [5376], []]" }, { "key": "aten::linear", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 793.8730000000005, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], [21504, 5376], []]" }, { "key": "comfy_kitchen::scaled_mm_nvfp4", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 646.9770000000003, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": -1024, "input_shapes": "[[37824, 2688], [21504, 2688], [], [], [37888, 336], [21504, 336], [], [], []]" }, { "key": "comfy_kitchen::rms_rope_split_half_", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 321.59999999999945, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128], [1, 37810, 56, 128], [1, 37810, 1, 48, 2, 2], [128], [128], [], []]" }, { "key": "aten::linear", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 231.9360000000006, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], [28672, 5376], []]" }, { "key": "aten::linear", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 212.3199999999997, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 7168], [5376, 7168], []]" }, { "key": "comfy_kitchen::scaled_mm_nvfp4", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 206.91200000000026, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": -1024, "input_shapes": "[[37824, 2688], [28672, 2688], [], [], [37888, 336], [28672, 336], [], [], []]" }, { "key": "aten::linear", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 200.52799999999934, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 14336], [5376, 14336], []]" }, { "key": "aten::contiguous", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 198.5599999999995, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[3, 5376], []]" }, { "key": "aten::empty", "count": 88, "device_type": "DeviceType.CPU", "cpu_time_total_us": 198.20800000000554, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 16019122176, "input_shapes": "[[], [], [], [], [], []]" }, { "key": "aten::clone", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 186.3679999999997, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[3, 5376], []]" }, { "key": "comfy_kitchen::scaled_mm_nvfp4", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 178.73599999999988, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": -1024, "input_shapes": "[[37824, 3584], [5376, 3584], [], [], [37888, 448], [5376, 448], [], [], []]" }, { "key": "comfy_kitchen::scaled_mm_nvfp4", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 177.96800000000076, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": -1024, "input_shapes": "[[37824, 7168], [5376, 7168], [], [], [37888, 896], [5376, 896], [], [], []]" }, { "key": "comfy_kitchen::quantize_nvfp4", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 135.51999999999998, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 7168], [], [], [], []]" }, { "key": "aten::zeros", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 128.43200000000184, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], [], [], [], []]" }, { "key": "aten::copy_", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 124.544000000001, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[3, 5376], [3, 5376], []]" }, { "key": "aten::mul", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 90.81600000000162, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 4096, "input_shapes": "[[], []]" }, { "key": "aten::mean", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 84.9280000000008, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 28672, "input_shapes": "[[1, 37810, 56, 128], [], [], []]" }, { "key": "sageattention_sm89::qk_int8_sv_f8_accum_f16_fuse_v_scale_attn_inst_buf", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 78.8159999999998, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128], [1, 37810, 56, 128], [1, 128, 56, 37824], [1, 37810, 56, 128], [1, 56, 1184], [1, 56, 591], [1, 56, 128], [], [], [], [], []]" }, { "key": "aten::_to_copy", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 73.10400000000027, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], [], [], [], [], [], []]" }, { "key": "aten::zero_", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 58.36799999999994, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 336]]" }, { "key": "aten::to", "count": 18, "device_type": "DeviceType.CPU", "cpu_time_total_us": 51.45600000000195, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], [], [], [], []]" }, { "key": "aten::copy_", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 45.904000000000906, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], [], []]" }, { "key": "aten::fill_", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 41.69599999999991, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 336], []]" }, { "key": "aten::to", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 33.58399999999983, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], [], [], [], [], []]" }, { "key": "aten::empty_like", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 31.839999999999918, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[3, 5376], [], [], [], [], []]" }, { "key": "aten::split", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 31.471999999999753, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 21504], [], []]" }, { "key": "aten::narrow", "count": 6, "device_type": "DeviceType.CPU", "cpu_time_total_us": 18.60799999999881, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 21504], [], [], []]" }, { "key": "aten::slice", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 16.5600000000004, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 21504], [], [], [], []]" }, { "key": "aten::zero_", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 15.487999999999829, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 448]]" }, { "key": "aten::reshape", "count": 10, "device_type": "DeviceType.CPU", "cpu_time_total_us": 14.78399999999965, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], []]" }, { "key": "aten::zero_", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 14.67199999999957, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 896]]" }, { "key": "aten::fill_", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 12.128000000000156, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 448], []]" }, { "key": "aten::empty_strided", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 12.064000000001215, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 2048, "input_shapes": "[[], [], [], [], [], []]" }, { "key": "aten::fill_", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 11.840000000000146, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 896], []]" }, { "key": "aten::squeeze", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 10.672000000000025, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 1, 56, 128], []]" }, { "key": "aten::view", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 10.544000000001688, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 336], []]" }, { "key": "aten::slice", "count": 6, "device_type": "DeviceType.CPU", "cpu_time_total_us": 9.920000000001437, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 21504], [], [], [], []]" }, { "key": "aten::slice", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 7.119999999999891, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 5376], [], [], [], []]" }, { "key": "aten::to", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 5.3119999999995855, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[5376], [], [], [], []]" }, { "key": "aten::view", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 4.688000000001466, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 7168], []]" }, { "key": "aten::view", "count": 10, "device_type": "DeviceType.CPU", "cpu_time_total_us": 4.623999999999796, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[], []]" }, { "key": "aten::as_strided", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 4.608000000000175, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 21504], [], [], []]" }, { "key": "aten::view", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 4.303999999999178, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810], []]" }, { "key": "aten::slice", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 4.079999999999927, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 28672], [], [], [], []]" }, { "key": "aten::as_strided", "count": 8, "device_type": "DeviceType.CPU", "cpu_time_total_us": 3.983999999999469, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128], [], [], []]" }, { "key": "aten::as_strided", "count": 6, "device_type": "DeviceType.CPU", "cpu_time_total_us": 3.2640000000001237, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 21504], [], [], []]" }, { "key": "aten::reshape", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 3.1359999999995125, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 7168], []]" }, { "key": "aten::reshape", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 3.087999999999738, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128], []]" }, { "key": "aten::reshape", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 2.207999999999629, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], []]" }, { "key": "aten::as_strided", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.7759999999989304, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 5376], [], [], []]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.7600000000002183, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[21504, 336], []]" }, { "key": "aten::alias", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.6640000000002146, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128]]" }, { "key": "aten::alias", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.551999999999225, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 21504]]" }, { "key": "aten::view", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.4399999999995998, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 896], []]" }, { "key": "aten::alias", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.3600000000014916, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376]]" }, { "key": "aten::view", "count": 4, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.360000000000582, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37888, 448], []]" }, { "key": "aten::as_strided", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 1.0080000000007203, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37824, 28672], [], [], []]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.975999999999658, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[28672, 336], []]" }, { "key": "aten::as_strided", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.8320000000003347, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 1, 56, 128], [], [], []]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.8160000000002583, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[1, 37810, 56, 128], []]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.7039999999997235, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 5376], []]" }, { "key": "aten::alias", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.6560000000008586, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[37810, 28672]]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.6080000000001746, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[5376, 896], []]" }, { "key": "aten::view", "count": 2, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.5600000000004002, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": 0, "input_shapes": "[[5376, 448], []]" }, { "key": "[memory]", "count": 74, "device_type": "DeviceType.CPU", "cpu_time_total_us": 0.0, "device_time_total_us": 0.0, "self_device_time_total_us": 0.0, "self_device_memory_usage_bytes": -16018849792, "input_shapes": "[]" } ] }