IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
863c0e6 verified
raw
history blame
8.2 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": "0.2.1",
"optimum_benchmark_commit": "d8c451da5aac8852b6a6e83b9b56a2aa6f007535",
"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": 955.645952,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.662485601999947,
"mean": 0.05096043092307285,
"stdev": 0.001101209430659164,
"p50": 0.05132497399998215,
"p90": 0.0520375321999893,
"p95": 0.0523162093999872,
"p99": 0.05257836187999146,
"values": [
0.05121247600001766,
0.051425492999982225,
0.04911277399997971,
0.05209774899998365,
0.05264389999999253,
0.048750125999987404,
0.05132497399998215,
0.05076386700000057,
0.0517966650000119,
0.050193400000011934,
0.05138464499998463,
0.05005046400000879,
0.05172906900000385
]
},
"throughput": {
"unit": "tokens/s",
"value": 39.24613594847919
},
"energy": {
"unit": "kWh",
"cpu": 1.7080156699470853e-06,
"ram": 7.137709808472189e-08,
"gpu": 0.0,
"total": 1.7793927680318072e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1123978.9415420673
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 955.645952,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.41250339700002314,
"mean": 0.031731030538463316,
"stdev": 0.0007622808331194323,
"p50": 0.032074269000020195,
"p90": 0.03241323679999368,
"p95": 0.032445448999988004,
"p99": 0.03246455059998425,
"values": [
0.03169805599998199,
0.032074269000020195,
0.032161913000010145,
0.03218302199999812,
0.0317578469999944,
0.03231234400001881,
0.03145086399999286,
0.0323480600000039,
0.03242953099999113,
0.03246932599998331,
0.0313359800000228,
0.03017547199999626,
0.030106713000009222
]
},
"throughput": {
"unit": "tokens/s",
"value": 31.514891985239267
},
"energy": {
"unit": "kWh",
"cpu": 9.987249389746722e-07,
"ram": 4.174241319631031e-08,
"gpu": 0.0,
"total": 1.040467352170983e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 961106.5622707468
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.9924372350000112,
"mean": 0.0827031029166676,
"stdev": 0.0012935412306620518,
"p50": 0.08273239200001115,
"p90": 0.08427710450000347,
"p95": 0.08437042434999284,
"p99": 0.08446047606998264,
"values": [
0.08352961800000003,
0.08126260400001684,
0.08427832600000329,
0.08448298899998008,
0.08106825999999501,
0.08275965800001472,
0.0831345679999913,
0.08426611100000514,
0.0826907990000052,
0.08270512600000757,
0.0802886419999993,
0.08197053399999277
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.091444755193878
},
"energy": null,
"efficiency": null
}
}
}