IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
9716a53 verified
raw
history blame
8.53 kB
{
"config": {
"name": "cpu_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.3.1+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",
"numactl": false,
"numactl_kwargs": {},
"start_method": "spawn"
},
"environment": {
"cpu": " AMD EPYC 7763 64-Core Processor",
"cpu_count": 4,
"cpu_ram_mb": 16757.338112,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-6.5.0-1022-azure-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.14",
"optimum_benchmark_version": "0.2.1",
"optimum_benchmark_commit": "074fc31a7e13e70facbebbf8b363a0c312f75f69",
"transformers_version": "4.42.3",
"transformers_commit": null,
"accelerate_version": "0.31.0",
"accelerate_commit": null,
"diffusers_version": "0.29.2",
"diffusers_commit": null,
"optimum_version": null,
"optimum_commit": null,
"timm_version": "1.0.7",
"timm_commit": null,
"peft_version": null,
"peft_commit": null
}
},
"report": {
"prefill": {
"memory": {
"unit": "MB",
"max_ram": 957.919232,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 15,
"total": 0.6209561489999089,
"mean": 0.04139707659999393,
"stdev": 0.001215984812479955,
"p50": 0.04136080299997502,
"p90": 0.04277917779999711,
"p95": 0.04362523439998824,
"p99": 0.04437743647999753,
"values": [
0.04322226899998327,
0.04136080299997502,
0.04158306799999423,
0.04160475899999483,
0.040270537999987255,
0.04112958199999639,
0.040301956999996946,
0.040833929999990914,
0.04099267699999132,
0.04141674699999953,
0.04188611499998274,
0.04211454100001788,
0.03962887000000137,
0.04456548699999985,
0.04004480599999738
]
},
"throughput": {
"unit": "tokens/s",
"value": 48.31259026634488
},
"energy": {
"unit": "kWh",
"cpu": 1.636626985337999e-06,
"ram": 6.839705853513481e-08,
"gpu": 0.0,
"total": 1.705024043873134e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1173003.9861823875
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 957.919232,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 15,
"total": 0.38539122700007056,
"mean": 0.02569274846667137,
"stdev": 0.0010644268041239298,
"p50": 0.025318746000010606,
"p90": 0.026641747400009307,
"p95": 0.027510887900010057,
"p99": 0.028760059180013398,
"values": [
0.025868852000002107,
0.025181119000023955,
0.025490657000005967,
0.025318746000010606,
0.0248253950000219,
0.025734653000000662,
0.02481574600000158,
0.025195935999988706,
0.025759087000011505,
0.024829101999984005,
0.02634183500001086,
0.02493068100000073,
0.026841689000008273,
0.029072352000014234,
0.025185376999985465
]
},
"throughput": {
"unit": "tokens/s",
"value": 38.921487955918764
},
"energy": {
"unit": "kWh",
"cpu": 1.0018698374430347e-06,
"ram": 4.186933423504592e-08,
"gpu": 0.0,
"total": 1.0437391716780795e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 958093.7720218381
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 14,
"total": 0.937497766000007,
"mean": 0.06696412614285764,
"stdev": 0.002035000141926469,
"p50": 0.06662766249999663,
"p90": 0.06783331170000224,
"p95": 0.07003707000000503,
"p99": 0.07295033879999976,
"values": [
0.06653081199999633,
0.06711912000000098,
0.06693043699999635,
0.06511442800001532,
0.06687713799999528,
0.06514225899999815,
0.06602905500000134,
0.06672451299999693,
0.06632880399999408,
0.06807621600000857,
0.06726653499998747,
0.06644808699999771,
0.07367865599999845,
0.06523170600002004
]
},
"throughput": {
"unit": "tokens/s",
"value": 14.933368918555797
},
"energy": null,
"efficiency": null
}
}
}