h3-blackwell-runtime/benchmarks/gb10-fused-block-residual-profile-component2-1344x768-124f-seed440420.json

1532 lines
47 KiB
JSON
Raw Normal View History

{
"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": 3,
"iterations": 5,
"hidden_shape": [
37810,
5376
],
"output_shape": [
37810,
5376
],
"segments": [
[
0,
100,
1
],
[
100,
514,
2
],
[
514,
37810,
0
]
],
"timings": {
"norm1": {
"count": 5,
"mean_s": 0.00370076420949772,
"p50_s": 0.0036445120349526405,
"p90_s": 0.003916181833483279,
"p95_s": 0.004001954442355782,
"p99_s": 0.004070572529453784,
"min_s": 0.00354526296723634,
"max_s": 0.004087727051228285
},
"modulate_msa": {
"count": 5,
"mean_s": 0.010961576015688479,
"p50_s": 0.010747502034064382,
"p90_s": 0.01159105401020497,
"p95_s": 0.011800702009350062,
"p99_s": 0.011968420408666134,
"min_s": 0.010485582984983921,
"max_s": 0.012010350008495152
},
"linear.attn_qkv_proj.flatten_contiguous": {
"count": 5,
"mean_s": 5.1711918786168095e-06,
"p50_s": 4.815985448658466e-06,
"p90_s": 6.588792894035578e-06,
"p95_s": 7.094396278262138e-06,
"p99_s": 7.4988789856433865e-06,
"min_s": 4.031986463814974e-06,
"max_s": 7.599999662488699e-06
},
"linear.attn_qkv_proj.pre_quant_scale": {
"count": 5,
"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": 5,
"mean_s": 1.361599424853921e-05,
"p50_s": 9.135983418673277e-06,
"p90_s": 2.324479864910245e-05,
"p95_s": 2.690240507945418e-05,
"p99_s": 2.982849022373557e-05,
"min_s": 8.032016921788454e-06,
"max_s": 3.056001150980592e-05
},
"linear.attn_qkv_proj.bias_cast": {
"count": 5,
"mean_s": 3.5776058211922647e-06,
"p50_s": 3.5199918784201145e-06,
"p90_s": 3.856001421809197e-06,
"p95_s": 3.9359903894364836e-06,
"p99_s": 3.999981563538312e-06,
"min_s": 3.296008799225092e-06,
"max_s": 4.01597935706377e-06
},
"linear.attn_qkv_proj.activation_scale": {
"count": 5,
"mean_s": 0.0017301246174611152,
"p50_s": 0.0017411670414730906,
"p90_s": 0.00177472640061751,
"p95_s": 0.0017832832061685621,
"p99_s": 0.0017901286506094038,
"min_s": 0.0016634880448691547,
"max_s": 0.0017918400117196143
},
"linear.attn_qkv_proj.scale_to_device": {
"count": 5,
"mean_s": 1.5584006905555726e-05,
"p50_s": 1.7023994587361813e-05,
"p90_s": 1.9052799325436352e-05,
"p95_s": 1.94304040633142e-05,
"p99_s": 1.9732487853616476e-05,
"min_s": 1.1184019967913628e-05,
"max_s": 1.9808008801192045e-05
},
"linear.attn_qkv_proj.activation_quant_pack": {
"count": 5,
"mean_s": 0.002322953403927386,
"p50_s": 0.0022722400026395917,
"p90_s": 0.002535033598542213,
"p95_s": 0.0025973887997679415,
"p99_s": 0.0026472729607485233,
"min_s": 0.0021412789938040078,
"max_s": 0.002659744000993669
},
"linear.attn_qkv_proj.activation_quant_wrap": {
"count": 5,
"mean_s": 6.976001895964146e-06,
"p50_s": 7.071997970342636e-06,
"p90_s": 7.951981388032436e-06,
"p95_s": 8.207978680729866e-06,
"p99_s": 8.412776514887811e-06,
"min_s": 6.0480087995529175e-06,
"max_s": 8.463975973427296e-06
},
"linear.attn_qkv_proj.gemm": {
"count": 5,
"mean_s": 0.03267693240195513,
"p50_s": 0.02632073196582496,
"p90_s": 0.04515579420840368,
"p95_s": 0.0506150125991553,
"p99_s": 0.054982387311756614,
"min_s": 0.025907867995556444,
"max_s": 0.05607423098990694
},
"linear.attn_qkv_proj.slice_reshape": {
"count": 5,
"mean_s": 6.319989915937186e-06,
"p50_s": 6.608024705201387e-06,
"p90_s": 7.007969543337822e-06,
"p95_s": 7.135968189686537e-06,
"p99_s": 7.2383671067655085e-06,
"min_s": 5.231995601207018e-06,
"max_s": 7.263966836035252e-06
},
"attn_qkv_proj": {
"count": 5,
"mean_s": 0.03682621520711109,
"p50_s": 0.030483131005894393,
"p90_s": 0.04947682461934165,
"p95_s": 0.0550504078157246,
"p99_s": 0.05950927437283098,
"min_s": 0.029794570989906788,
"max_s": 0.06062399101210758
},
"attn_qkv_split_view": {
"count": 5,
"mean_s": 1.826880034059286e-05,
"p50_s": 2.0159990526735783e-05,
"p90_s": 2.2012810222804547e-05,
"p95_s": 2.2102415096014737e-05,
"p99_s": 2.217409899458289e-05,
"min_s": 1.3535958714783192e-05,
"max_s": 2.219201996922493e-05
},
"attn_qk_rms_rope": {
"count": 5,
"mean_s": 0.012522334605455399,
"p50_s": 0.01253089401870966,
"p90_s": 0.01304144160822034,
"p95_s": 0.013201063813176005,
"p99_s": 0.01332876157714054,
"min_s": 0.011644751008134335,
"max_s": 0.013360686018131673
},
"attn_q_transpose_contiguous": {
"count": 5,
"mean_s": 0.010179851192515344,
"p50_s": 0.00496414297958836,
"p90_s": 0.020518390822689983,
"p95_s": 0.02548829743172973,
"p99_s": 0.029464222718961538,
"min_s": 0.004930094990413636,
"max_s": 0.030458204040769488
},
"attn_k_transpose_contiguous": {
"count": 5,
"mean_s": 0.010719105391763151,
"p50_s": 0.004932318988721818,
"p90_s": 0.022215538995806132,
"p95_s": 0.027844894991721953,
"p99_s": 0.032348379788454625,
"min_s": 0.004929247021209449,
"max_s": 0.03347425098763779
},
"attn_v_transpose_contiguous": {
"count": 5,
"mean_s": 0.008646603603847326,
"p50_s": 0.005328848026692867,
"p90_s": 0.01569883298361674,
"p95_s": 0.01915314298821613,
"p99_s": 0.021916590991895645,
"min_s": 0.004975487012416124,
"max_s": 0.022607452992815524
},
"attention_kernel": {
"count": 5,
"mean_s": 0.26568923201411965,
"p50_s": 0.2572870340081863,
"p90_s": 0.28179349241545426,
"p95_s": 0.28840694121317934,
"p99_s": 0.2936977002513595,
"min_s": 0.2569375950261019,
"max_s": 0.2950203900109045
},
"attn_output_reshape": {
"count": 5,
"mean_s": 0.004784690390806645,
"p50_s": 0.004718751995824277,
"p90_s": 0.004893067781813443,
"p95_s": 0.004894261376466601,
"p99_s": 0.00489521625218913,
"min_s": 0.004707070998847485,
"max_s": 0.0048954549711197615
},
"linear.attn_out_proj.flatten_contiguous": {
"count": 5,
"mean_s": 4.870397970080376e-06,
"p50_s": 4.5120250433683395e-06,
"p90_s": 5.759997293353081e-06,
"p95_s": 5.999987479299307e-06,
"p99_s": 6.1919796280562874e-06,
"min_s": 4.239962436258793e-06,
"max_s": 6.239977665245533e-06
},
"linear.attn_out_proj.pre_quant_scale": {
"count": 5,
"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": 5,
"mean_s": 1.5878397971391677e-05,
"p50_s": 1.0415969882160425e-05,
"p90_s": 2.6598421391099693e-05,
"p95_s": 3.005122998729348e-05,
"p99_s": 3.2813476864248514e-05,
"min_s": 9.215960744768381e-06,
"max_s": 3.350403858348727e-05
},
"linear.attn_out_proj.bias_cast": {
"count": 5,
"mean_s": 3.6063836887478827e-06,
"p50_s": 3.439956344664097e-06,
"p90_s": 3.920029848814011e-06,
"p95_s": 4.032033029943705e-06,
"p99_s": 4.121635574847459e-06,
"min_s": 3.4239492379128933e-06,
"max_s": 4.144036211073399e-06
},
"linear.attn_out_proj.activation_scale": {
"count": 5,
"mean_s": 0.0022977533983066676,
"p50_s": 0.002296319988090545,
"p90_s": 0.0023102719802409414,
"p95_s": 0.0023118719807825984,
"p99_s": 0.002313151981215924,
"min_s": 0.002284880029037595,
"max_s": 0.0023134719813242555
},
"linear.attn_out_proj.scale_to_device": {
"count": 5,
"mean_s": 1.2838398106396198e-05,
"p50_s": 1.1263997294008732e-05,
"p90_s": 1.6198388766497374e-05,
"p95_s": 1.7027172725647687e-05,
"p99_s": 1.769019989296794e-05,
"min_s": 1.0271964129060507e-05,
"max_s": 1.7855956684798002e-05
},
"linear.attn_out_proj.activation_quant_pack": {
"count": 5,
"mean_s": 0.002899941592477262,
"p50_s": 0.002914238953962922,
"p90_s": 0.002939288597553968,
"p95_s": 0.002945931791327894,
"p99_s": 0.002951246346347034,
"min_s": 0.002855567028746009,
"max_s": 0.002952574985101819
},
"linear.attn_out_proj.activation_quant_wrap": {
"count": 5,
"mean_s": 7.366400677710772e-06,
"p50_s": 6.735965143889189e-06,
"p90_s": 8.963211439549922e-06,
"p95_s": 9.4336224719882e-06,
"p99_s": 9.809951297938823e-06,
"min_s": 6.176007445901632e-06,
"max_s": 9.90403350442648e-06
},
"linear.attn_out_proj.gemm": {
"count": 5,
"mean_s": 0.008071339805610478,
"p50_s": 0.007845567015465349,
"p90_s": 0.008501621428877115,
"p95_s": 0.008666610228829086,
"p99_s": 0.008798601268790661,
"min_s": 0.00783475098432973,
"max_s": 0.008831599028781056
},
"linear.attn_out_proj.slice_reshape": {
"count": 5,
"mean_s": 5.833595059812069e-06,
"p50_s": 5.615991540253162e-06,
"p90_s": 6.400002166628838e-06,
"p95_s": 6.608001422137022e-06,
"p99_s": 6.77440082654357e-06,
"min_s": 5.424022674560547e-06,
"max_s": 6.8160006776452065e-06
},
"attn_out_proj": {
"count": 5,
"mean_s": 0.013364516384899616,
"p50_s": 0.013153581996448338,
"p90_s": 0.01383550518658012,
"p95_s": 0.013990103593096138,
"p99_s": 0.014113782318308949,
"min_s": 0.013065422012005001,
"max_s": 0.014144701999612153
},
"gate_msa": {
"count": 5,
"mean_s": 0.008748206798918546,
"p50_s": 0.008709535002708435,
"p90_s": 0.008930900797713548,
"p95_s": 0.008934161404613405,
"p99_s": 0.00893676989013329,
"min_s": 0.008583375019952655,
"max_s": 0.008937422011513263
},
"norm2": {
"count": 5,
"mean_s": 0.0035379068111069502,
"p50_s": 0.0035068950382992625,
"p90_s": 0.003613071620929986,
"p95_s": 0.0036341758328489963,
"p99_s": 0.003651059202384204,
"min_s": 0.003484688000753522,
"max_s": 0.0036552800447680056
},
"modulate_mlp": {
"count": 5,
"mean_s": 0.010690554999746383,
"p50_s": 0.010643246001563966,
"p90_s": 0.011026267206761985,
"p95_s": 0.011032692599110307,
"p99_s": 0.011037832912988961,
"min_s": 0.010254477965645492,
"max_s": 0.011039117991458625
},
"linear.mlp_fc1.flatten_contiguous": {
"count": 5,
"mean_s": 4.796788562089205e-06,
"p50_s": 4.767964128404856e-06,
"p90_s": 5.372788291424513e-06,
"p95_s": 5.574396345764399e-06,
"p99_s": 5.735682789236307e-06,
"min_s": 4.127970896661282e-06,
"max_s": 5.776004400104284e-06
},
"linear.mlp_fc1.pre_quant_scale": {
"count": 5,
"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": 5,
"mean_s": 8.927995804697274e-06,
"p50_s": 8.736038580536842e-06,
"p90_s": 9.878387209028005e-06,
"p95_s": 1.0147178545594216e-05,
"p99_s": 1.0362211614847183e-05,
"min_s": 7.96798849478364e-06,
"max_s": 1.0415969882160425e-05
},
"linear.mlp_fc1.bias_cast": {
"count": 5,
"mean_s": 3.6192010156810285e-06,
"p50_s": 3.6800047382712364e-06,
"p90_s": 3.6992132663726807e-06,
"p95_s": 3.705616109073162e-06,
"p99_s": 3.710738383233547e-06,
"min_s": 3.375986125320196e-06,
"max_s": 3.7120189517736435e-06
},
"linear.mlp_fc1.activation_scale": {
"count": 5,
"mean_s": 0.0017399997916072607,
"p50_s": 0.0017451830208301544,
"p90_s": 0.0017586175817996264,
"p95_s": 0.0017593887750990689,
"p99_s": 0.0017600057297386229,
"min_s": 0.001707071962300688,
"max_s": 0.0017601599683985114
},
"linear.mlp_fc1.scale_to_device": {
"count": 5,
"mean_s": 1.1795188765972852e-05,
"p50_s": 1.0351999662816525e-05,
"p90_s": 1.6096001490950585e-05,
"p95_s": 1.7888005822896956e-05,
"p99_s": 1.9321609288454058e-05,
"min_s": 9.007984772324562e-06,
"max_s": 1.968001015484333e-05
},
"linear.mlp_fc1.activation_quant_pack": {
"count": 5,
"mean_s": 0.0022628605947829783,
"p50_s": 0.0021891199867241085,
"p90_s": 0.002411609189584851,
"p95_s": 0.002436860592570156,
"p99_s": 0.0024570617149583993,
"min_s": 0.002152992004994303,
"max_s": 0.0024621119955554605
},
"linear.mlp_fc1.activation_quant_wrap": {
"count": 5,
"mean_s": 6.5312138758599755e-06,
"p50_s": 6.560003384947777e-06,
"p90_s": 6.992032285779714e-06,
"p95_s": 7.072032894939184e-06,
"p99_s": 7.13603338226676e-06,
"min_s": 6.095971912145615e-06,
"max_s": 7.152033504098654e-06
},
"linear.mlp_fc1.gemm": {
"count": 5,
"mean_s": 0.0716201048111543,
"p50_s": 0.07167574996128678,
"p90_s": 0.07178817201638595,
"p95_s": 0.07180112900678068,
"p99_s": 0.07181149459909647,
"min_s": 0.07133280503330752,
"max_s": 0.07181408599717543
},
"linear.mlp_fc1.slice_reshape": {
"count": 5,
"mean_s": 5.792011506855488e-06,
"p50_s": 5.6800199672579765e-06,
"p90_s": 6.160023622214794e-06,
"p95_s": 6.192026194185018e-06,
"p99_s": 6.217628251761198e-06,
"min_s": 5.4559786804020405e-06,
"max_s": 6.224028766155243e-06
},
"mlp_fc1": {
"count": 5,
"mean_s": 0.07571263879071921,
"p50_s": 0.07571077399188653,
"p90_s": 0.07579628340899944,
"p95_s": 0.07580330419586971,
"p99_s": 0.07580892082536593,
"min_s": 0.07558842096477747,
"max_s": 0.07581032498273998
},
"mlp_swiglu": {
"count": 5,
"mean_s": 0.024767270009033383,
"p50_s": 0.024613836023490876,
"p90_s": 0.025068361801095305,
"p95_s": 0.025185571413021534,
"p99_s": 0.025279339102562515,
"min_s": 0.024597676005214453,
"max_s": 0.025302781024947762
},
"linear.mlp_fc2.flatten_contiguous": {
"count": 5,
"mean_s": 4.336016718298197e-06,
"p50_s": 4.096014890819788e-06,
"p90_s": 4.9184192903339865e-06,
"p95_s": 5.123228766024112e-06,
"p99_s": 5.287076346576214e-06,
"min_s": 3.936002030968666e-06,
"max_s": 5.328038241714239e-06
},
"linear.mlp_fc2.pre_quant_scale": {
"count": 5,
"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": 5,
"mean_s": 8.10560304671526e-06,
"p50_s": 7.872004061937332e-06,
"p90_s": 8.53120582178235e-06,
"p95_s": 8.553615771234035e-06,
"p99_s": 8.571543730795383e-06,
"min_s": 7.760012522339821e-06,
"max_s": 8.57602572068572e-06
},
"linear.mlp_fc2.bias_cast": {
"count": 5,
"mean_s": 3.6800047382712364e-06,
"p50_s": 3.503984771668911e-06,
"p90_s": 4.086398985236883e-06,
"p95_s": 4.259194247424602e-06,
"p99_s": 4.397430457174778e-06,
"min_s": 3.4240074455738068e-06,
"max_s": 4.431989509612322e-06
},
"linear.mlp_fc2.activation_scale": {
"count": 5,
"mean_s": 0.004378968803212047,
"p50_s": 0.004356223973445594,
"p90_s": 0.0044430166250094775,
"p95_s": 0.004468475829344243,
"p99_s": 0.0044888431928120556,
"min_s": 0.004333471006248146,
"max_s": 0.0044939350336790085
},
"linear.mlp_fc2.scale_to_device": {
"count": 5,
"mean_s": 9.596801828593015e-06,
"p50_s": 9.232026059180498e-06,
"p90_s": 1.0777614079415798e-05,
"p95_s": 1.1180818546563387e-05,
"p99_s": 1.1503382120281458e-05,
"min_s": 8.751987479627132e-06,
"max_s": 1.1584023013710976e-05
},
"linear.mlp_fc2.activation_quant_pack": {
"count": 5,
"mean_s": 0.0058938198140822354,
"p50_s": 0.0058150230324827135,
"p90_s": 0.006084485398605466,
"p95_s": 0.006107090192381292,
"p99_s": 0.006125174027401954,
"min_s": 0.005752607015892863,
"max_s": 0.006129694986157119
},
"linear.mlp_fc2.activation_quant_wrap": {
"count": 5,
"mean_s": 6.204796954989434e-06,
"p50_s": 6.191956344991922e-06,
"p90_s": 6.3615967519581315e-06,
"p95_s": 6.396800745278597e-06,
"p99_s": 6.4249639399349686e-06,
"min_s": 6.0480087995529175e-06,
"max_s": 6.432004738599062e-06
},
"linear.mlp_fc2.gemm": {
"count": 5,
"mean_s": 0.053530597186181694,
"p50_s": 0.05350962496595457,
"p90_s": 0.05387938078492879,
"p95_s": 0.05388907039305195,
"p99_s": 0.053896822079550474,
"min_s": 0.0530537529848516,
"max_s": 0.053898760001175106
},
"linear.mlp_fc2.slice_reshape": {
"count": 5,
"mean_s": 5.484803114086389e-06,
"p50_s": 5.4879928939044476e-06,
"p90_s": 5.6319870054721825e-06,
"p95_s": 5.679985042661429e-06,
"p99_s": 5.718383472412825e-06,
"min_s": 5.328038241714239e-06,
"max_s": 5.727983079850674e-06
},
"mlp_fc2": {
"count": 5,
"mean_s": 0.06388071699766443,
"p50_s": 0.06395071098813787,
"p90_s": 0.06436533080413936,
"p95_s": 0.0644813884049654,
"p99_s": 0.06457423448562623,
"min_s": 0.06331373500870541,
"max_s": 0.06459744600579143
},
"gate_mlp": {
"count": 5,
"mean_s": 0.011803047789726406,
"p50_s": 0.009327662992291152,
"p90_s": 0.017453504574950787,
"p95_s": 0.02013837477425113,
"p99_s": 0.02228627093369141,
"min_s": 0.008719118020962924,
"max_s": 0.022823244973551482
},
"block_total": {
"count": 5,
"mean_s": 0.5766410392010585,
"p50_s": 0.5561805270262994,
"p90_s": 0.6259831115952693,
"p95_s": 0.6395874358015136,
"p99_s": 0.6504708951665089,
"min_s": 0.543343793018721,
"max_s": 0.6531917600077577
}
},
"module_forward": {
"count": 5,
"mean_s": 0.5217330385930836,
"p50_s": 0.5216766290250234,
"p90_s": 0.5225755279883743,
"p95_s": 0.5228595179971307,
"p99_s": 0.523086710004136,
"min_s": 0.5204838289646432,
"max_s": 0.5231435080058873
},
"module_forward_checksum": 333782528.0,
"fused_elementwise": true,
"profiler_summary": {
"profiled_iterations": 2,
"runtime_kernel_launches": 0,
"runtime_kernel_launches_per_block": 0.0,
"positive_self_device_allocated_bytes": 22523446272
},
"profiler_top_events": [
{
"key": "aten::rms_norm",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1042.591999999999,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": -606208,
"input_shapes": "[[37810, 5376], [], [5376], []]"
},
{
"key": "comfy_kitchen::quantize_nvfp4",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1023.7440000000001,
"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::_fused_rms_norm",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1021.6320000000005,
"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": 644.1119999999996,
"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": "aten::zeros",
"count": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 518.3519999999994,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[], [], [], [], []]"
},
{
"key": "comfy_kitchen::scaled_mm_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 499.7760000000003,
"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": 465.85599999999977,
"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::zero_",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 430.86400000000094,
"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::contiguous",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 420.71900000000005,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], []]"
},
{
"key": "aten::clone",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 409.96700000000055,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], []]"
},
{
"key": "aten::_to_copy",
"count": 16,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 297.3599999999992,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[], [], [], [], [], [], []]"
},
{
"key": "aten::linear",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 257.03999999999905,
"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": "aten::empty",
"count": 76,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 220.49600000000498,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 20354950144,
"input_shapes": "[[], [], [], [], [], []]"
},
{
"key": "aten::linear",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 219.36000000000058,
"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": "comfy_kitchen::scaled_mm_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 216.04799999999977,
"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": "aten::linear",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 209.45600000000013,
"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::to",
"count": 24,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 203.599999999999,
"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": 16,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 199.64800000000014,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[], [], []]"
},
{
"key": "comfy_kitchen::scaled_mm_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 192.84800000000087,
"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::scaled_mm_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 180.46399999999994,
"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::to",
"count": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 119.04000000000133,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[], [], [], [], [], []]"
},
{
"key": "comfy_kitchen::quantize_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 116.27199999999993,
"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::mul",
"count": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 111.32800000000043,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 4096,
"input_shapes": "[[], []]"
},
{
"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": 108.91200000000026,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], [1, 56, 37810, 128], [1, 56, 128, 37824], [1, 56, 37810, 128], [1, 56, 1184], [1, 56, 591], [1, 56, 128], [], [], [], [], []]"
},
{
"key": "aten::mean",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 104.86400000000049,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 28672,
"input_shapes": "[[1, 56, 37810, 128], [], [], []]"
},
{
"key": "comfy_kitchen::quantize_nvfp4",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 93.5039999999999,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[37810, 14336], [], [], [], []]"
},
{
"key": "aten::copy_",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 77.1359999999986,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], [1, 56, 37810, 128], []]"
},
{
"key": "aten::silu",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 74.30399999999827,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 2168455168,
"input_shapes": "[[37810, 14336]]"
},
{
"key": "aten::fill_",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 54.400000000001,
"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::mul_",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 49.34399999999914,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[37810, 14336], [37810, 14336]]"
},
{
"key": "aten::empty_strided",
"count": 16,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 49.072000000001935,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 8192,
"input_shapes": "[[], [], [], [], [], []]"
},
{
"key": "aten::reshape",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 47.35999999999967,
"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::split",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 38.35199999999941,
"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::empty_like",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 28.943999999999505,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], [], [], [], [], []]"
},
{
"key": "aten::clone",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 26.35200000000077,
"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::narrow",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 22.192000000000007,
"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::zero_",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 20.54400000000078,
"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::chunk",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 20.43199999999979,
"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::reshape",
"count": 16,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 19.968000000000302,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[], []]"
},
{
"key": "aten::transpose",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 18.064000000000306,
"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::slice",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 17.871999999999844,
"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": 17.391999999999825,
"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::copy_",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 16.31999999999971,
"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], []]"
},
{
"key": "aten::fill_",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 15.984000000000378,
"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": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 15.423999999998841,
"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::split",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 15.072000000001935,
"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::fill_",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 14.495999999998276,
"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::narrow",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 10.720000000000255,
"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::squeeze",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 10.256000000000313,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 1, 128], []]"
},
{
"key": "aten::slice",
"count": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 9.95199999999977,
"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": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 9.791999999999007,
"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": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 8.752000000001317,
"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::empty_like",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 6.11200000000008,
"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::to",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 6.0479999999990355,
"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": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 6.000000000000455,
"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": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 5.839999999999236,
"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::as_strided",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 5.584000000000287,
"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": 16,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 5.359999999999218,
"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": 14,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 5.104000000001179,
"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": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 4.191999999999098,
"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::slice",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 4.11200000000008,
"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": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 3.967999999999847,
"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": 6,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 3.936000000001968,
"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": 8,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 3.8239999999998417,
"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::transpose",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 3.4559999999992215,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], [], []]"
},
{
"key": "aten::reshape",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 3.119999999999891,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[37810, 14336], []]"
},
{
"key": "aten::reshape",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 2.768000000000029,
"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": 2.4799999999995634,
"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::reshape",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 2.2880000000004657,
"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::_unsafe_view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 2.256000000000313,
"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": 2.1279999999997017,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128]]"
},
{
"key": "aten::alias",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 2.063999999999851,
"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::as_strided",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.6800000000012005,
"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::as_strided",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.359999999998763,
"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::alias",
"count": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.295999999999367,
"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": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.2479999999995925,
"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": 4,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.1359999999976935,
"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::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 1.087999999999738,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[37810, 14336], []]"
},
{
"key": "aten::as_strided",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.8159999999998035,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 37810, 128], [], [], []]"
},
{
"key": "aten::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.7999999999997272,
"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::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.7520000000004075,
"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::as_strided",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.7520000000004075,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[1, 56, 1, 128], [], [], []]"
},
{
"key": "aten::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.7200000000011642,
"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::alias",
"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, 28672]]"
},
{
"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": "[[28672, 336], []]"
},
{
"key": "aten::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.543999999999869,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[5376, 448], []]"
},
{
"key": "aten::view",
"count": 2,
"device_type": "DeviceType.CPU",
"cpu_time_total_us": 0.5119999999997162,
"device_time_total_us": 0.0,
"self_device_time_total_us": 0.0,
"self_device_memory_usage_bytes": 0,
"input_shapes": "[[5376, 896], []]"
},
{
"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": -22522835968,
"input_shapes": "[]"
}
]
}