IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
b2f6a0a 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": 16757.3504,
"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": "6422f8f56820540ec4876445962a50a44754b92e",
"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": 956.645376,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.6462017299999161,
"mean": 0.04970782538460893,
"stdev": 0.0020330109769887933,
"p50": 0.05068381199998839,
"p90": 0.05136708300001373,
"p95": 0.05155970220001791,
"p99": 0.051636767640018205,
"values": [
0.050683911999954034,
0.05165603400001828,
0.05078850899997178,
0.05068381199998839,
0.05149548100001766,
0.04755527599996867,
0.050116962999993575,
0.0506895319999785,
0.04847438699999884,
0.04869581499997366,
0.05085349099999803,
0.050571160000004056,
0.04393735800005061
]
},
"throughput": {
"unit": "tokens/s",
"value": 40.235113576689706
},
"energy": {
"unit": "kWh",
"cpu": 1.7085316676448509e-06,
"ram": 7.140149890809109e-08,
"gpu": 0.0,
"total": 1.779933166552942e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1123637.6947080796
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 956.645376,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.40297509000004084,
"mean": 0.030998083846156987,
"stdev": 0.0013468495296399831,
"p50": 0.031345950000002176,
"p90": 0.031951286400033044,
"p95": 0.03219337080003015,
"p99": 0.03236183736001976,
"values": [
0.032403954000017166,
0.031544504000009965,
0.03205298200003881,
0.031345950000002176,
0.03131585300002371,
0.031220865000022968,
0.03144647899995334,
0.03111336200004189,
0.031524687000000995,
0.03145440499997676,
0.030714631000023473,
0.030126922999954786,
0.0267104949999748
]
},
"throughput": {
"unit": "tokens/s",
"value": 32.26005855597348
},
"energy": {
"unit": "kWh",
"cpu": 1.0148321027341093e-06,
"ram": 4.241213541862194e-08,
"gpu": 0.0,
"total": 1.057244238152732e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 945855.237525104
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.9665089690000173,
"mean": 0.08054241408333478,
"stdev": 0.003224147982260754,
"p50": 0.08157314600003929,
"p90": 0.08285569790000977,
"p95": 0.08303225564999366,
"p99": 0.08320171592997895,
"values": [
0.08324408099997527,
0.08282648300001938,
0.08206126099997846,
0.0828589440000087,
0.07877808500001038,
0.08155088900002738,
0.08184023499995874,
0.08000393399998984,
0.08019483399999672,
0.08159540300005119,
0.08086701099995253,
0.07068780900004867
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.415818564431538
},
"energy": null,
"efficiency": null
}
}
}