IlyasMoutawwakil's picture
Upload cuda_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
d5c5ade verified
raw
history blame
13.8 kB
{
"config": {
"name": "cuda_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.3.0+cu121",
"_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": true,
"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 7R32",
"cpu_count": 16,
"cpu_ram_mb": 66697.29792,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-5.10.215-203.850.amzn2.x86_64-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.12",
"gpu": [
"NVIDIA A10G"
],
"gpu_count": 1,
"gpu_vram_mb": 24146608128,
"optimum_benchmark_version": "0.2.0",
"optimum_benchmark_commit": null,
"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": 1071.140864,
"max_global_vram": 1365.77024,
"max_process_vram": 0.0,
"max_reserved": 725.614592,
"max_allocated": 522.007552
},
"latency": {
"unit": "s",
"count": 56,
"total": 0.4978674545288085,
"mean": 0.00889049025944301,
"stdev": 0.0003923566450490379,
"p50": 0.008917056083679199,
"p90": 0.009441648006439209,
"p95": 0.009564608097076417,
"p99": 0.009740710735321045,
"values": [
0.00957804775238037,
0.00930191993713379,
0.009199295997619629,
0.008995391845703126,
0.008966752052307129,
0.009777600288391114,
0.009328415870666504,
0.009464703559875488,
0.00928825569152832,
0.00941859245300293,
0.00901084804534912,
0.008912639617919922,
0.008967904090881348,
0.008826463699340821,
0.008918399810791016,
0.008932512283325195,
0.009067647933959961,
0.009046976089477538,
0.008861408233642579,
0.009034496307373047,
0.0089619197845459,
0.008915712356567383,
0.009039135932922363,
0.008887904167175293,
0.008851136207580566,
0.008923935890197753,
0.009131168365478516,
0.009710528373718262,
0.009338560104370118,
0.009492575645446777,
0.008717151641845703,
0.009168383598327636,
0.008887807846069335,
0.009167776107788086,
0.009560128211975098,
0.00903059196472168,
0.00877228832244873,
0.008801823616027831,
0.008684479713439941,
0.008444767951965332,
0.008430239677429199,
0.008419648170471191,
0.008426591873168946,
0.008458815574645995,
0.008392064094543458,
0.008388319969177245,
0.008385087966918945,
0.008389951705932618,
0.008397664070129394,
0.008394975662231446,
0.008394368171691894,
0.00837929630279541,
0.008389920234680175,
0.008399328231811524,
0.008411999702453613,
0.00842313575744629
]
},
"throughput": {
"unit": "tokens/s",
"value": 224.959472609028
},
"energy": {
"unit": "kWh",
"cpu": 1.057174899393604e-07,
"ram": 5.7796648949144064e-08,
"gpu": 1.8470252871428628e-07,
"total": 3.4821666760279074e-07
},
"efficiency": {
"unit": "tokens/kWh",
"value": 5743550.4560091635
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 1071.140864,
"max_global_vram": 1365.77024,
"max_process_vram": 0.0,
"max_reserved": 725.614592,
"max_allocated": 522.4576
},
"latency": {
"unit": "s",
"count": 56,
"total": 0.5151266231536864,
"mean": 0.00919868969917297,
"stdev": 0.00040756636637087417,
"p50": 0.009235791683197022,
"p90": 0.009674848079681397,
"p95": 0.009740551948547364,
"p99": 0.010056236934661867,
"values": [
0.00952291202545166,
0.009675488471984864,
0.0094234561920166,
0.009526432037353516,
0.009653599739074707,
0.00967420768737793,
0.009664992332458495,
0.009775808334350586,
0.009495295524597169,
0.00972879981994629,
0.009368320465087891,
0.009280159950256347,
0.00944223976135254,
0.009565631866455078,
0.009375935554504394,
0.009312159538269043,
0.009367648124694825,
0.009217056274414062,
0.009299455642700195,
0.00927235221862793,
0.009183808326721192,
0.009205632209777832,
0.009169471740722657,
0.009222240447998046,
0.0092391357421875,
0.009404512405395507,
0.010306719779968262,
0.00972486400604248,
0.009610719680786133,
0.009555456161499023,
0.009118047714233399,
0.009232447624206544,
0.0092871675491333,
0.009604063987731934,
0.009851296424865723,
0.009162495613098145,
0.00901478385925293,
0.009043744087219238,
0.00896070384979248,
0.008731840133666992,
0.00879196834564209,
0.008673824310302735,
0.00865657615661621,
0.008648896217346191,
0.00866592025756836,
0.00866703987121582,
0.008674015998840332,
0.008653887748718262,
0.008679615974426269,
0.00868518352508545,
0.008682175636291504,
0.008646240234375,
0.00864185619354248,
0.008659263610839844,
0.008757696151733399,
0.00867136001586914
]
},
"throughput": {
"unit": "tokens/s",
"value": 108.71113524895912
},
"energy": {
"unit": "kWh",
"cpu": 9.835689284262199e-08,
"ram": 5.377790440415142e-08,
"gpu": 1.7194526618226395e-07,
"total": 3.2408006342903737e-07
},
"efficiency": {
"unit": "tokens/kWh",
"value": 3085657.2583304443
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 55,
"total": 0.9957550086975098,
"mean": 0.018104636521772903,
"stdev": 0.000779112071230643,
"p50": 0.018223104476928712,
"p90": 0.019157811737060545,
"p95": 0.019348889350891112,
"p99": 0.019601572341918946,
"values": [
0.01902796745300293,
0.01865318489074707,
0.018589696884155273,
0.018553855895996094,
0.019579904556274414,
0.01901568031311035,
0.0192542724609375,
0.01883545684814453,
0.019183616638183593,
0.01843609619140625,
0.01821900749206543,
0.01841663932800293,
0.018421760559082033,
0.018358272552490236,
0.018250751495361327,
0.018480127334594726,
0.018313215255737304,
0.01820159912109375,
0.018313215255737304,
0.018223104476928712,
0.018149375915527344,
0.018241535186767577,
0.018136064529418947,
0.018114559173583983,
0.018348031997680665,
0.019297279357910157,
0.01962700843811035,
0.018977792739868164,
0.019119104385375976,
0.017870847702026366,
0.018506784439086914,
0.018190303802490235,
0.018711551666259766,
0.01946931266784668,
0.018335744857788085,
0.01782681655883789,
0.017869823455810546,
0.01765990447998047,
0.01719910430908203,
0.017275903701782228,
0.017122367858886718,
0.01710892868041992,
0.017125375747680666,
0.017085439682006837,
0.017075199127197266,
0.017076223373413087,
0.017089536666870117,
0.017101823806762697,
0.017105920791625977,
0.017093631744384767,
0.017060863494873048,
0.017056768417358398,
0.017084415435791016,
0.01719398307800293,
0.017120256423950195
]
},
"throughput": {
"unit": "tokens/s",
"value": 55.23446984408581
},
"energy": null,
"efficiency": null
}
}
}