IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
126d8ab verified
raw
history blame
8.33 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": "b92e3e6e00670fa1351d9aef1be896254e5f33e0",
"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": 945.62304,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.63180773600007,
"mean": 0.045129124000005,
"stdev": 0.0006773019204614954,
"p50": 0.04508576349999771,
"p90": 0.04584225179999919,
"p95": 0.046137948349999645,
"p99": 0.04652379927000368,
"values": [
0.046620262000004686,
0.04458333900001321,
0.04463016200000425,
0.04443067200000428,
0.045279906000018855,
0.04396656600002302,
0.045153609999999844,
0.044558249000004935,
0.045323927999987745,
0.04587824099999693,
0.045758277000004455,
0.04501791699999558,
0.044858679000014945,
0.04574792799999727
]
},
"throughput": {
"unit": "tokens/s",
"value": 44.31727945793449
},
"energy": {
"unit": "kWh",
"cpu": 1.730579281774696e-06,
"ram": 7.23231251243345e-08,
"gpu": 0.0,
"total": 1.8029024068990306e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1109322.3861406757
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 945.62304,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.3855475449999517,
"mean": 0.027539110357139407,
"stdev": 0.0010498280970777697,
"p50": 0.027215351500004203,
"p90": 0.02813615549999895,
"p95": 0.029281207849992084,
"p99": 0.030691900769992057,
"values": [
0.027102476000010256,
0.02701685599998882,
0.026924448999977812,
0.027206404999986944,
0.02740718900000161,
0.02684920899997678,
0.02709805300000312,
0.027581373999993275,
0.027679878000014924,
0.027224298000021463,
0.02739818300000252,
0.028331702999992103,
0.031044573999992053,
0.02668289799999002
]
},
"throughput": {
"unit": "tokens/s",
"value": 36.311993634927056
},
"energy": {
"unit": "kWh",
"cpu": 1.021381914135723e-06,
"ram": 4.2686825832303366e-08,
"gpu": 0.0,
"total": 1.0640687399680258e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 939788.9087786274
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 13,
"total": 0.9438911560000065,
"mean": 0.0726070120000005,
"stdev": 0.0012046050905438965,
"p50": 0.0724765520000119,
"p90": 0.07333238259998894,
"p95": 0.07439264599998978,
"p99": 0.07561652839999851,
"values": [
0.07161263800000484,
0.07158011800001418,
0.07165658299999222,
0.07272466500000974,
0.07080854099999101,
0.07227428499999178,
0.07214889100001187,
0.07303028499998732,
0.0731124179999938,
0.07317093700001465,
0.0733727439999825,
0.0759224990000007,
0.0724765520000119
]
},
"throughput": {
"unit": "tokens/s",
"value": 13.772774453244173
},
"energy": null,
"efficiency": null
}
}
}