IlyasMoutawwakil's picture
Upload cuda_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
f919336 verified
raw
history blame
13.2 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": "0.2.0",
"optimum_benchmark_commit": "14fc8ac8cd252f949be10158bdcefa6a3b7c596a",
"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": 1048.977408,
"max_global_vram": 2098.7904,
"max_process_vram": 325403.820032,
"max_reserved": 725.614592,
"max_allocated": 512.832512
},
"latency": {
"unit": "s",
"count": 56,
"total": 0.503788423538208,
"mean": 0.008996221848896572,
"stdev": 0.0007065452092312092,
"p50": 0.008924727439880372,
"p90": 0.00942976760864258,
"p95": 0.010141928911209106,
"p99": 0.011910649919509892,
"values": [
0.01133392906188965,
0.010119688987731934,
0.010208648681640625,
0.00999280834197998,
0.009517127990722657,
0.0093424072265625,
0.008410567283630372,
0.008466727256774902,
0.008873447418212891,
0.008810727119445801,
0.008806407928466797,
0.012615530967712403,
0.009094087600708008,
0.009039688110351562,
0.008929766654968262,
0.008882086753845215,
0.008924647331237793,
0.008923208236694336,
0.008902567863464356,
0.008896488189697266,
0.008719207763671875,
0.008901608467102052,
0.00892480754852295,
0.00891824722290039,
0.0088508882522583,
0.008357927322387696,
0.008658886909484863,
0.008972806930541992,
0.008940966606140136,
0.008681767463684082,
0.008536487579345704,
0.008363046646118165,
0.008932647705078125,
0.008327366828918457,
0.008302566528320312,
0.00869728660583496,
0.008980486869812012,
0.008932168006896973,
0.00891264820098877,
0.008941448211669922,
0.00892912769317627,
0.008947687149047851,
0.0089320068359375,
0.008964167594909669,
0.008960647583007813,
0.009045767784118652,
0.008953766822814941,
0.008957766532897949,
0.00897072696685791,
0.00896176815032959,
0.008915047645568847,
0.009007047653198243,
0.008448966979980469,
0.008701127052307129,
0.008120647430419922,
0.008028326988220215
]
},
"throughput": {
"unit": "tokens/s",
"value": 222.31554908189702
},
"energy": null,
"efficiency": null
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 1048.977408,
"max_global_vram": 2098.7904,
"max_process_vram": 305801.191424,
"max_reserved": 725.614592,
"max_allocated": 513.056256
},
"latency": {
"unit": "s",
"count": 56,
"total": 0.5047401008605958,
"mean": 0.009013216086796352,
"stdev": 0.00044162054033490355,
"p50": 0.009030087471008301,
"p90": 0.009449128150939942,
"p95": 0.009985568523406981,
"p99": 0.010404592609405518,
"values": [
0.010168329238891601,
0.010534567832946777,
0.01029824924468994,
0.009924648284912109,
0.009522248268127441,
0.008973287582397462,
0.008455527305603028,
0.008751527786254883,
0.008684806823730469,
0.008951207160949708,
0.008799527168273926,
0.009617447853088379,
0.009198246955871582,
0.009061767578125,
0.009049448013305664,
0.00904432773590088,
0.00903984832763672,
0.0090651273727417,
0.009012967109680177,
0.009034087181091308,
0.009048486709594726,
0.009027207374572754,
0.009028488159179688,
0.008736487388610839,
0.008774247169494629,
0.008620648384094239,
0.009376008033752442,
0.00902896785736084,
0.008821288108825683,
0.008531046867370605,
0.008431846618652344,
0.008997286796569824,
0.009037607192993164,
0.008537768363952637,
0.008835846900939942,
0.00889040756225586,
0.009072327613830566,
0.009009126663208009,
0.009038726806640624,
0.009051206588745118,
0.009167207717895507,
0.009068967819213868,
0.009076807975769043,
0.009069287300109863,
0.009079687118530274,
0.008988167762756348,
0.009050888061523437,
0.009018088340759277,
0.009053607940673829,
0.009056007385253906,
0.009031207084655761,
0.008791207313537597,
0.008555848121643067,
0.00843632698059082,
0.00812656593322754,
0.008088006019592284
]
},
"throughput": {
"unit": "tokens/s",
"value": 110.94818878967307
},
"energy": null,
"efficiency": null
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 55,
"total": 0.9892410240173343,
"mean": 0.0179862004366788,
"stdev": 0.0009813673698406225,
"p50": 0.01798209571838379,
"p90": 0.01838235206604004,
"p95": 0.020219007682800293,
"p99": 0.021368254737854003,
"values": [
0.020591856002807617,
0.020596336364746094,
0.020059215545654298,
0.019155696868896486,
0.01837937545776367,
0.01697633361816406,
0.017233774185180663,
0.017590415954589843,
0.017840496063232424,
0.017613616943359375,
0.022274419784545897,
0.018384336471557617,
0.01813729667663574,
0.018004335403442383,
0.017972496032714842,
0.018000335693359376,
0.018022415161132812,
0.017952335357666015,
0.01796321678161621,
0.017796335220336915,
0.017966575622558593,
0.017983535766601562,
0.017694095611572265,
0.017690895080566406,
0.01697377395629883,
0.01806465530395508,
0.01804033660888672,
0.0178382568359375,
0.017215373992919922,
0.016986574172973633,
0.017374576568603514,
0.018019695281982422,
0.016920494079589843,
0.017184816360473632,
0.01758065605163574,
0.018085296630859376,
0.017970256805419922,
0.01798209571838379,
0.01802497673034668,
0.018131376266479492,
0.01805409622192383,
0.018046255111694335,
0.01806449508666992,
0.0180740966796875,
0.018072175979614257,
0.01804017639160156,
0.01801201629638672,
0.01805681610107422,
0.018056175231933595,
0.017978096008300783,
0.017832496643066405,
0.017040334701538085,
0.017219696044921875,
0.01627313232421875,
0.016148012161254884
]
},
"throughput": {
"unit": "tokens/s",
"value": 55.59817947768031
},
"energy": null,
"efficiency": null
}
}
}