IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
e98d5fb verified
raw
history blame
9.52 kB
{
"config": {
"name": "cpu_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.4.0+cpu",
"_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
"task": "text-generation",
"library": "transformers",
"model_type": "gpt2",
"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": {},
"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,
"memory": true,
"latency": 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": 16757.342208,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-6.5.0-1025-azure-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.14",
"optimum_benchmark_version": "0.4.0",
"optimum_benchmark_commit": "e4c497aec760cceef111d142b57bbf79ef01f3fe",
"transformers_version": "4.44.2",
"transformers_commit": null,
"accelerate_version": "0.33.0",
"accelerate_commit": null,
"diffusers_version": "0.30.1",
"diffusers_commit": null,
"optimum_version": null,
"optimum_commit": null,
"timm_version": "1.0.9",
"timm_commit": null,
"peft_version": null,
"peft_commit": null
}
},
"report": {
"load": {
"memory": {
"unit": "MB",
"max_ram": 1111.30624,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 1,
"total": 4.840812504000013,
"mean": 4.840812504000013,
"stdev": 0.0,
"p50": 4.840812504000013,
"p90": 4.840812504000013,
"p95": 4.840812504000013,
"p99": 4.840812504000013,
"values": [
4.840812504000013
]
},
"throughput": null,
"energy": {
"unit": "kWh",
"cpu": 6.831305662222242e-05,
"ram": 2.855311511575416e-06,
"gpu": 0,
"total": 7.116836813379784e-05
},
"efficiency": null
},
"prefill": {
"memory": {
"unit": "MB",
"max_ram": 964.665344,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.6514608420000343,
"mean": 0.046532917285716735,
"stdev": 0.0011893933845824529,
"p50": 0.04661327249999658,
"p90": 0.04763813200000868,
"p95": 0.04843660715000908,
"p99": 0.04933134463000499,
"values": [
0.04718022000000133,
0.045830832999996574,
0.04585134200002017,
0.04677872199999911,
0.046661953000011636,
0.04656459199998153,
0.04565858199998729,
0.04584101300000043,
0.046917139000015595,
0.046828454000007014,
0.04575968200001057,
0.04955502900000397,
0.047834380000011834,
0.044198900999987245
]
},
"throughput": {
"unit": "tokens/s",
"value": 42.9803269741246
},
"energy": {
"unit": "kWh",
"cpu": 1.7000295446860133e-06,
"ram": 7.10468526279848e-08,
"gpu": 0.0,
"total": 1.7710763973139982e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1129256.7633068713
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 965.451776,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 14,
"total": 0.3934041239999715,
"mean": 0.028100294571426536,
"stdev": 0.0007619635158025801,
"p50": 0.02818939199998738,
"p90": 0.028840471999996,
"p95": 0.029010672850002094,
"p99": 0.02922316097000163,
"values": [
0.028505515999995623,
0.02851165800001354,
0.02877705199998104,
0.029276283000001513,
0.027944189000010056,
0.02886765200000241,
0.02773636200001306,
0.028119215000003805,
0.02823815799999352,
0.02814062599998124,
0.027854741999988164,
0.028253616000000648,
0.02607059499999309,
0.027108459999993784
]
},
"throughput": {
"unit": "tokens/s",
"value": 35.58681555661835
},
"energy": {
"unit": "kWh",
"cpu": 1.0086162060546754e-06,
"ram": 4.215238178330891e-08,
"gpu": 0.0,
"total": 1.0507685878379843e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 951684.3304742831
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 14,
"total": 0.38778311199988025,
"mean": 0.02769879371427716,
"stdev": 0.0007510585224946331,
"p50": 0.027736721999986003,
"p90": 0.028451626599996872,
"p95": 0.028627257399995186,
"p99": 0.028843392279987938,
"values": [
0.028087115999994694,
0.02812799299999824,
0.028381263999989415,
0.028897425999986126,
0.027532821999983526,
0.028481782000000067,
0.027341044999985797,
0.02772400799997854,
0.02784650699999247,
0.027749435999993466,
0.02746059699998682,
0.027686168000002453,
0.025726072999987082,
0.02674087500000155
]
},
"throughput": {
"unit": "tokens/s",
"value": 36.10265523890149
},
"energy": null,
"efficiency": null
}
}
}