IlyasMoutawwakil's picture
Upload cuda_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
c8650ce verified
raw
history blame
13.9 kB
{
"config": {
"name": "cuda_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.2.2+rocm5.7",
"_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
"task": "text-generation",
"library": "transformers",
"model": "openai-community/gpt2",
"processor": "openai-community/gpt2",
"device": "cuda",
"device_ids": "0",
"seed": 42,
"inter_op_num_threads": null,
"intra_op_num_threads": null,
"model_kwargs": {},
"processor_kwargs": {},
"hub_kwargs": {},
"no_weights": true,
"device_map": null,
"torch_dtype": null,
"eval_mode": true,
"to_bettertransformer": false,
"low_cpu_mem_usage": null,
"attn_implementation": null,
"cache_implementation": null,
"autocast_enabled": false,
"autocast_dtype": null,
"torch_compile": false,
"torch_compile_target": "forward",
"torch_compile_config": {},
"quantization_scheme": null,
"quantization_config": {},
"deepspeed_inference": false,
"deepspeed_inference_config": {},
"peft_type": null,
"peft_config": {}
},
"scenario": {
"name": "inference",
"_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
"iterations": 1,
"duration": 1,
"warmup_runs": 1,
"input_shapes": {
"batch_size": 1,
"num_choices": 2,
"sequence_length": 2
},
"new_tokens": null,
"latency": true,
"memory": true,
"energy": false,
"forward_kwargs": {},
"generate_kwargs": {
"max_new_tokens": 2,
"min_new_tokens": 2
},
"call_kwargs": {
"num_inference_steps": 2
}
},
"launcher": {
"name": "process",
"_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
"device_isolation": true,
"device_isolation_action": "error",
"start_method": "spawn"
},
"environment": {
"cpu": " AMD EPYC 7763 64-Core Processor",
"cpu_count": 128,
"cpu_ram_mb": 1082015.236096,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-5.15.0-84-generic-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.12",
"gpu": [
"Advanced Micro Devices, Inc. [AMD/ATI]"
],
"gpu_count": 1,
"gpu_vram_mb": 68702699520,
"optimum_benchmark_version": "0.2.1",
"optimum_benchmark_commit": "56d026bf244c6516d8cb780280ce7cc6505f270e",
"transformers_version": "4.40.2",
"transformers_commit": null,
"accelerate_version": "0.30.1",
"accelerate_commit": null,
"diffusers_version": "0.27.2",
"diffusers_commit": null,
"optimum_version": null,
"optimum_commit": null,
"timm_version": "1.0.3",
"timm_commit": null,
"peft_version": null,
"peft_commit": null
}
},
"report": {
"prefill": {
"memory": {
"unit": "MB",
"max_ram": 1048.86272,
"max_global_vram": 2098.76992,
"max_process_vram": 281290.838016,
"max_reserved": 725.614592,
"max_allocated": 512.832512
},
"latency": {
"unit": "s",
"count": 61,
"total": 0.4988777332305908,
"mean": 0.008178323495583456,
"stdev": 0.0005084790264905845,
"p50": 0.007969289779663087,
"p90": 0.008714250564575195,
"p95": 0.009018891334533692,
"p99": 0.009997772407531736,
"values": [
0.011285134315490723,
0.008870091438293457,
0.008714250564575195,
0.009139531135559082,
0.008409130096435546,
0.00823760986328125,
0.008188810348510742,
0.008135689735412597,
0.008166729927062988,
0.008205769538879394,
0.008143850326538086,
0.008111049652099609,
0.009126091957092285,
0.008149290084838867,
0.007977610111236572,
0.007953289031982422,
0.007969289779663087,
0.007976970195770264,
0.007940330028533936,
0.00792992877960205,
0.007978569984436036,
0.007952970027923584,
0.007945449829101562,
0.008135370254516602,
0.007964489936828613,
0.007928648948669433,
0.007967209815979004,
0.007977769851684571,
0.00796432876586914,
0.00799089002609253,
0.007967050075531007,
0.007937129974365234,
0.00793248987197876,
0.007935688972473145,
0.00793297004699707,
0.00794832992553711,
0.0079457688331604,
0.007939529895782471,
0.00795857000350952,
0.007930409908294678,
0.007912489891052246,
0.007965610027313233,
0.007911049842834472,
0.00791248893737793,
0.007945610046386718,
0.007993450164794922,
0.00791825008392334,
0.00887937068939209,
0.009018891334533692,
0.008451690673828125,
0.00848529052734375,
0.00839841079711914,
0.008320650100708009,
0.008151049613952637,
0.007988649845123291,
0.00809457015991211,
0.00794416904449463,
0.007910570144653321,
0.007953450202941895,
0.007940649032592773,
0.007917290210723876
]
},
"throughput": {
"unit": "tokens/s",
"value": 244.54889820389974
},
"energy": null,
"efficiency": null
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 1048.86272,
"max_global_vram": 2098.76992,
"max_process_vram": 267569.373184,
"max_reserved": 725.614592,
"max_allocated": 513.056256
},
"latency": {
"unit": "s",
"count": 61,
"total": 0.5017876567840575,
"mean": 0.008226027160394386,
"stdev": 0.00042518155981721684,
"p50": 0.008045129776000977,
"p90": 0.008636171340942383,
"p95": 0.008928010940551758,
"p99": 0.009816171646118163,
"values": [
0.009205930709838868,
0.008959211349487305,
0.00881153106689453,
0.008631690979003907,
0.008367530822753906,
0.008286250114440919,
0.008177289962768555,
0.008133609771728516,
0.00826737117767334,
0.008146570205688477,
0.008116649627685548,
0.010731533050537109,
0.00821249008178711,
0.008062089920043945,
0.008060648918151855,
0.008051850318908692,
0.008032329559326172,
0.008045129776000977,
0.008030570030212403,
0.008038889884948731,
0.008018730163574218,
0.008259849548339844,
0.00812240982055664,
0.008017129898071289,
0.008007209777832031,
0.008034729957580566,
0.008636171340942383,
0.008047369956970215,
0.00803553009033203,
0.00817809009552002,
0.008020170211791992,
0.00803345012664795,
0.008037928581237794,
0.008003529548645019,
0.008033129692077636,
0.007997769832611084,
0.008027850151062011,
0.00801984977722168,
0.008028969764709473,
0.008006889343261719,
0.008150250434875488,
0.008009129524230958,
0.008013449668884277,
0.007980169773101807,
0.007983850002288819,
0.008001290321350097,
0.008012969017028809,
0.008734890937805176,
0.008928010940551758,
0.008504170417785645,
0.008498889923095703,
0.008421930313110351,
0.008328330993652345,
0.008055689811706544,
0.00826097011566162,
0.007991849899291992,
0.007980169773101807,
0.008004810333251953,
0.008020970344543457,
0.007984329223632812,
0.007985610008239746
]
},
"throughput": {
"unit": "tokens/s",
"value": 121.56536569860491
},
"energy": null,
"efficiency": null
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 60,
"total": 0.9822864904403684,
"mean": 0.016371441507339473,
"stdev": 0.0006170808170715157,
"p50": 0.0160540189743042,
"p90": 0.017585045623779296,
"p95": 0.017898886489868163,
"p99": 0.018152883090972897,
"values": [
0.01777650260925293,
0.017600021362304687,
0.017892982482910155,
0.016875539779663086,
0.01657714080810547,
0.016434581756591796,
0.01631298065185547,
0.016438739776611328,
0.01642546081542969,
0.016304500579833986,
0.01824898338317871,
0.01808610153198242,
0.01624578094482422,
0.016058900833129882,
0.01604129981994629,
0.016033939361572266,
0.016052978515625,
0.016003379821777342,
0.015994739532470704,
0.01603218078613281,
0.01623858070373535,
0.0160550594329834,
0.016225940704345704,
0.016005300521850587,
0.01599153995513916,
0.016629779815673828,
0.01605825996398926,
0.016027700424194335,
0.016198099136352537,
0.016019859313964845,
0.015984498977661132,
0.016012500762939454,
0.015973139762878417,
0.015991378784179687,
0.015973299980163575,
0.016003379821777342,
0.015990260124206544,
0.01601473808288574,
0.01596417999267578,
0.016090740203857423,
0.01600305938720703,
0.015954898834228515,
0.015923060417175294,
0.01596273994445801,
0.016022739410400392,
0.015962260246276855,
0.01758338165283203,
0.018011062622070313,
0.017011379241943358,
0.01695553970336914,
0.016844980239868163,
0.016746740341186523,
0.016237300872802736,
0.016278099060058594,
0.016116819381713866,
0.01595506000518799,
0.015945459365844728,
0.016003219604492188,
0.015954419136047362,
0.01592929935455322
]
},
"throughput": {
"unit": "tokens/s",
"value": 61.081976168787
},
"energy": null,
"efficiency": null
}
}
}