IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
cccc92b verified
raw
history blame
8.21 kB
{
"config": {
"name": "cpu_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.3.0+cpu",
"_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
"task": "text-generation",
"library": "transformers",
"model": "openai-community/gpt2",
"processor": "openai-community/gpt2",
"device": "cpu",
"device_ids": null,
"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": false,
"device_isolation_action": "error",
"start_method": "spawn"
},
"environment": {
"cpu": " AMD EPYC 7763 64-Core Processor",
"cpu_count": 4,
"cpu_ram_mb": 16757.354496,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-6.5.0-1021-azure-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.14",
"optimum_benchmark_version": "0.2.0",
"optimum_benchmark_commit": "e9058512717b423b9cdf21d451280e352f72e99c",
"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": 956.8256,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.6219809429999827,
"mean": 0.0478446879230756,
"stdev": 0.000811586685782954,
"p50": 0.048065213999962,
"p90": 0.04858335019998776,
"p95": 0.048702704599998016,
"p99": 0.048810608120008966,
"values": [
0.04804461600002696,
0.048465610999983255,
0.04861278499998889,
0.0488375840000117,
0.04836251899996569,
0.045606254000006174,
0.047167433000026904,
0.047654250000050524,
0.047280934000013985,
0.0482910160000074,
0.048126067999987754,
0.047466658999951505,
0.048065213999962
]
},
"throughput": {
"unit": "tokens/s",
"value": 41.80192382518176
},
"energy": {
"unit": "kWh",
"cpu": 1.6345091440059522e-06,
"ram": 6.830680039653221e-08,
"gpu": 0.0,
"total": 1.7028159444024843e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1174525.060429709
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 956.8256,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.39072484399991936,
"mean": 0.03005575723076303,
"stdev": 0.0004925865733221267,
"p50": 0.03009674800000539,
"p90": 0.030735641800026768,
"p95": 0.030882855400000152,
"p99": 0.030980324679962905,
"values": [
0.029936168999995516,
0.030471685000009074,
0.03024234899999101,
0.030145797999978186,
0.029653461999998854,
0.030801631000031193,
0.03009674800000539,
0.02937488299994584,
0.03002595499998506,
0.031004691999953593,
0.030148359000008895,
0.029392190000010032,
0.029430923000006715
]
},
"throughput": {
"unit": "tokens/s",
"value": 33.271495784390616
},
"energy": {
"unit": "kWh",
"cpu": 9.785618384679161e-07,
"ram": 4.089793152646829e-08,
"gpu": 0.0,
"total": 1.0194597699943845e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 980911.6842398875
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.9348950980000268,
"mean": 0.07790792483333557,
"stdev": 0.0009168054790122415,
"p50": 0.07781592550000482,
"p90": 0.07902765240002622,
"p95": 0.07911032395001313,
"p99": 0.07918034159000228,
"values": [
0.07892808900004411,
0.07882882499995958,
0.07903871500002424,
0.07811028700001543,
0.07634774299998526,
0.07728256500001862,
0.07710626599998704,
0.0773395499999765,
0.07919784599999957,
0.07833176299999423,
0.076861885000028,
0.07752156399999421
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.835664691868622
},
"energy": null,
"efficiency": null
}
}
}