IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
86fd9b7 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": "0.2.0",
"optimum_benchmark_commit": "553a573935b71390502b7c324bca2f06f9fbc412",
"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": 946.413568,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.6467838759998585,
"mean": 0.04975260584614297,
"stdev": 0.002905657029950617,
"p50": 0.051133996000032766,
"p90": 0.05275953199999321,
"p95": 0.05290272219999679,
"p99": 0.0530197524399955,
"values": [
0.051498493999986295,
0.05304900999999518,
0.051133996000032766,
0.05257687199997463,
0.05176727499997469,
0.0492591569999945,
0.051979368999980124,
0.05280519699999786,
0.04960489000001189,
0.04471888699998772,
0.04739713099996834,
0.04514574999996057,
0.045847847999993974
]
},
"throughput": {
"unit": "tokens/s",
"value": 40.198899454329755
},
"energy": {
"unit": "kWh",
"cpu": 1.7257139878572476e-06,
"ram": 7.21190328254149e-08,
"gpu": 0.0,
"total": 1.7978330206826625e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1112450.3649624656
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 946.413568,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.3989463730001148,
"mean": 0.03068818253847037,
"stdev": 0.0019340917355519804,
"p50": 0.03185052800000676,
"p90": 0.03259064619999208,
"p95": 0.03267772399999558,
"p99": 0.03269836399999804,
"values": [
0.03270352399999865,
0.03231113499998628,
0.03185052800000676,
0.03266052399999353,
0.0316733870000121,
0.031924345000049925,
0.0321337540000286,
0.03212823400002662,
0.02856189699997458,
0.027912098000001606,
0.027338931000031153,
0.028837561000045753,
0.028910454999959256
]
},
"throughput": {
"unit": "tokens/s",
"value": 32.58583328440552
},
"energy": {
"unit": "kWh",
"cpu": 1.0752917199895003e-06,
"ram": 4.493357244351722e-08,
"gpu": 0.0,
"total": 1.1202252924330176e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 892677.5772292195
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.9619339139999852,
"mean": 0.08016115949999876,
"stdev": 0.004725155067696069,
"p50": 0.08208317050002734,
"p90": 0.08519843409995928,
"p95": 0.08530386924996662,
"p99": 0.08538075704998334,
"values": [
0.08539997899998752,
0.08300659400003951,
0.08522523399994952,
0.08353937600003292,
0.08115974700001516,
0.08414191699995399,
0.08495723500004715,
0.07827982699996028,
0.07269390100003648,
0.07476389400000016,
0.07400990999997248,
0.07475629999999
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.474869453454144
},
"energy": null,
"efficiency": null
}
}
}