IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
87499d3 verified
raw
history blame
8.32 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": "2db7e171d23768cdeb341987283e17e0b2db9bd9",
"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": 955.86304,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.6466785789999676,
"mean": 0.04619132707142626,
"stdev": 0.002444838147601616,
"p50": 0.04648068749997947,
"p90": 0.048878021100006205,
"p95": 0.0489982560000044,
"p99": 0.04913566079999328,
"values": [
0.04803570899997567,
0.04881326899999294,
0.0464519340000038,
0.04890577200001189,
0.04805475400002024,
0.04607904900001358,
0.04238081900001589,
0.04396958400002404,
0.04250237600001583,
0.04271998199999416,
0.04452450899998439,
0.04650944099995513,
0.048561368999969545,
0.04917001199999049
]
},
"throughput": {
"unit": "tokens/s",
"value": 43.298171470747604
},
"energy": {
"unit": "kWh",
"cpu": 1.71323047168013e-06,
"ram": 7.159635701472294e-08,
"gpu": 0.0,
"total": 1.784826828694853e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1120556.8898034166
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 955.86304,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.39997615100008943,
"mean": 0.02856972507143496,
"stdev": 0.002029412536081166,
"p50": 0.02900631450000901,
"p90": 0.03077352510003948,
"p95": 0.030807604550020073,
"p99": 0.030833569709997732,
"values": [
0.030136748000018088,
0.03079012800003511,
0.030840060999992147,
0.030668350999974336,
0.03025345699995796,
0.026768334999985655,
0.0261342329999934,
0.0260023760000081,
0.026297046000024693,
0.026190408000047682,
0.026894841000000724,
0.03073478500004967,
0.03038950100000193,
0.027875880999999936
]
},
"throughput": {
"unit": "tokens/s",
"value": 35.00208691191908
},
"energy": {
"unit": "kWh",
"cpu": 9.313993599871692e-07,
"ram": 3.892871889679738e-08,
"gpu": 0.0,
"total": 9.703280788839672e-07
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1030579.2667055046
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 13,
"total": 0.968865634999986,
"mean": 0.0745281257692297,
"stdev": 0.004295798324842498,
"p50": 0.07706898899999715,
"p90": 0.07948535479998782,
"p95": 0.07963069979999773,
"p99": 0.07965194556000597,
"values": [
0.07965725700000803,
0.07725985500002253,
0.07961299499999086,
0.0782805489999987,
0.07306524000000536,
0.06852501099996289,
0.06997215900003084,
0.06879245999999739,
0.06896410000001651,
0.07142139299998007,
0.07706898899999715,
0.07897479399997565,
0.07727083300000004
]
},
"throughput": {
"unit": "tokens/s",
"value": 13.417753226431843
},
"energy": null,
"efficiency": null
}
}
}