IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
3c95e15 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": "2516ce57a5b64eefeb78dc75a171e0cdff88823e",
"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": 956.510208,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.6357577669999728,
"mean": 0.04890444361538252,
"stdev": 0.0008820830605062864,
"p50": 0.04921229699999685,
"p90": 0.04982055379998655,
"p95": 0.049928289999991195,
"p99": 0.050039097999995195,
"values": [
0.05006679999999619,
0.04983594999998786,
0.04832375000000866,
0.04955616799998097,
0.04920199800000091,
0.04694990999999504,
0.049388175000018464,
0.04921639500000197,
0.047933985000014445,
0.04808931499999858,
0.04921229699999685,
0.048224054999991495,
0.049758968999981334
]
},
"throughput": {
"unit": "tokens/s",
"value": 40.89607921376934
},
"energy": {
"unit": "kWh",
"cpu": 1.694125656920355e-06,
"ram": 7.079968758763816e-08,
"gpu": 0.0,
"total": 1.764925344507993e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1133192.4073862392
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 956.510208,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 13,
"total": 0.39826846500002944,
"mean": 0.030636035769233034,
"stdev": 0.0003253756376968252,
"p50": 0.030561998999985462,
"p90": 0.03104573480001136,
"p95": 0.031231302600008348,
"p99": 0.031380361320007065,
"values": [
0.03068658300000493,
0.03037712600001896,
0.03141762600000675,
0.031107087000009415,
0.03019482600001311,
0.03044065499997828,
0.030740413999978955,
0.030752866999989692,
0.030392584999987093,
0.030474567999988267,
0.03080032600001914,
0.030321803000049385,
0.030561998999985462
]
},
"throughput": {
"unit": "tokens/s",
"value": 32.64129887863213
},
"energy": {
"unit": "kWh",
"cpu": 1.0548160118373317e-06,
"ram": 4.408370176471442e-08,
"gpu": 0.0,
"total": 1.0988997136020455e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 910001.1471675923
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 12,
"total": 0.953466211999995,
"mean": 0.07945551766666625,
"stdev": 0.0009842037005582196,
"p50": 0.07985784400001705,
"p90": 0.0803253735000169,
"p95": 0.08050314630000628,
"p99": 0.08067198925999207,
"values": [
0.08027954099998169,
0.07970219400002065,
0.08071419999998852,
0.07941578000000504,
0.07734190499999727,
0.08015944600001035,
0.0800676959999862,
0.07834084600000324,
0.0785156829999778,
0.08001349400001345,
0.07858496099999002,
0.0803304660000208
]
},
"throughput": {
"unit": "tokens/s",
"value": 12.585658357865398
},
"energy": null,
"efficiency": null
}
}
}