IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
c396591 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": "0.2.1",
"optimum_benchmark_commit": "56d026bf244c6516d8cb780280ce7cc6505f270e",
"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": 956.57984,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.6370267230000195,
"mean": 0.04900205561538611,
"stdev": 0.0015196532018327974,
"p50": 0.04874450999997748,
"p90": 0.05058189000000653,
"p95": 0.05151805420000528,
"p99": 0.05236686123999788,
"values": [
0.05081071500001144,
0.04961812000001942,
0.0468596620000028,
0.047920792000013535,
0.049416053000015836,
0.04647974399998134,
0.048309767999995756,
0.048697683000000325,
0.04864558500000271,
0.04874450999997748,
0.04927843800001597,
0.049666589999986854,
0.05257906299999604
]
},
"throughput": {
"unit": "tokens/s",
"value": 40.814614303078784
},
"energy": {
"unit": "kWh",
"cpu": 1.6467359882813915e-06,
"ram": 6.881930964928765e-08,
"gpu": 0.0,
"total": 1.7155552979306792e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1165803.2838769003
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 956.57984,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.4022535950000474,
"mean": 0.03094258423077288,
"stdev": 0.0011829603900231369,
"p50": 0.030767002000004595,
"p90": 0.03173786039999982,
"p95": 0.032875080400003746,
"p99": 0.034212845680008284,
"values": [
0.031760275999999976,
0.03080419200000506,
0.030259606000015538,
0.03081651599998736,
0.03089553199998818,
0.03062072000000171,
0.030287067000017487,
0.029965026000013495,
0.02990729900000133,
0.030767002000004595,
0.029974874000004093,
0.031648197999999184,
0.03454728700000942
]
},
"throughput": {
"unit": "tokens/s",
"value": 32.31792123572809
},
"energy": {
"unit": "kWh",
"cpu": 1.0072342885865103e-06,
"ram": 4.209258276205929e-08,
"gpu": 0.0,
"total": 1.049326871348569e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 952991.8915683774
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.9568434859999968,
"mean": 0.0797369571666664,
"stdev": 0.002517996565532965,
"p50": 0.07901532250001253,
"p90": 0.0812344256999836,
"p95": 0.08391430074999703,
"p99": 0.08644860175001015,
"values": [
0.08044263899998327,
0.07712888700001486,
0.07876194400000713,
0.08036543499997606,
0.07708972400001812,
0.07862227199998983,
0.07869105200001059,
0.07857551499998294,
0.07949273799999901,
0.07926870100001793,
0.08132240199998364,
0.08708217700001342
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.541236028229637
},
"energy": null,
"efficiency": null
}
}
}