IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
8f109b2 verified
raw
history blame
8.34 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": 16757.3504,
"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": "99a3ca8898b9c4f2a5e074afab82f89e395d4214",
"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": 955.916288,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.6623793490001049,
"mean": 0.04731281064286463,
"stdev": 0.0017174148638615624,
"p50": 0.046850515500011625,
"p90": 0.049434636199981695,
"p95": 0.049660089750008754,
"p99": 0.04999368275003405,
"values": [
0.045801589999996395,
0.04927965599995332,
0.04667580000000271,
0.04606988000000456,
0.045568115000037324,
0.04723987200003421,
0.04696534900000415,
0.04464512399999876,
0.048946384000032594,
0.04550677000003134,
0.046735682000019096,
0.05007708100004038,
0.04943248999995831,
0.04943555599999172
]
},
"throughput": {
"unit": "tokens/s",
"value": 42.271849269255476
},
"energy": {
"unit": "kWh",
"cpu": 1.7904939073504822e-06,
"ram": 7.482568369020942e-08,
"gpu": 0.0,
"total": 1.8653195910406916e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1072202.323723072
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 955.916288,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.401089769999885,
"mean": 0.02864926928570607,
"stdev": 0.001259856833549332,
"p50": 0.02798316699997372,
"p90": 0.030162063099982105,
"p95": 0.030686828299974423,
"p99": 0.03130535685998211,
"values": [
0.027583842000012737,
0.02990901800001211,
0.027710718999969686,
0.027801678000002994,
0.027511426999978994,
0.027814882999962265,
0.02918781300002138,
0.027417060000004767,
0.027288089999956355,
0.028151450999985173,
0.031459988999984034,
0.029653370999994877,
0.029329918000030375,
0.030270510999969247
]
},
"throughput": {
"unit": "tokens/s",
"value": 34.904904206367604
},
"energy": {
"unit": "kWh",
"cpu": 1.041590324555985e-06,
"ram": 4.353105560736285e-08,
"gpu": 0.0,
"total": 1.0851213801633474e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 921555.890687055
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 13,
"total": 0.9903058000000442,
"mean": 0.07617736923077263,
"stdev": 0.0025798671397240258,
"p50": 0.07615217100004656,
"p90": 0.07956659520001494,
"p95": 0.07969771340002581,
"p99": 0.07971720428003891,
"values": [
0.0791070920000152,
0.07450163300001122,
0.07383535999997548,
0.0731522770000197,
0.07508697499997652,
0.07615217100004656,
0.07207004899998992,
0.07628368600001068,
0.07366509299998825,
0.07816508399997701,
0.07968147100001488,
0.07888283199997659,
0.07972207700004219
]
},
"throughput": {
"unit": "tokens/s",
"value": 13.127258267092266
},
"energy": null,
"efficiency": null
}
}
}