IlyasMoutawwakil's picture
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
f398c28 verified
raw
history blame
8.46 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": "c1d0b062e90b79e7705510c58cea731c0d90da8a",
"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": 955.994112,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 15,
"total": 0.6500200730000358,
"mean": 0.04333467153333572,
"stdev": 0.0008503176008886599,
"p50": 0.04321825499999932,
"p90": 0.04396499099999005,
"p95": 0.04478463509998676,
"p99": 0.045756328619998496,
"values": [
0.04348617400000876,
0.045999252000001434,
0.04284785400000146,
0.04321825499999932,
0.0429543820000049,
0.04279751000001397,
0.04351635000000442,
0.04345768100000669,
0.04324075599998878,
0.042828436999997166,
0.043501452000015206,
0.043110293000012234,
0.04252533199999675,
0.04426408499998047,
0.04227226000000428
]
},
"throughput": {
"unit": "tokens/s",
"value": 46.15242089608262
},
"energy": {
"unit": "kWh",
"cpu": 1.6414997202378734e-06,
"ram": 6.860053029377866e-08,
"gpu": 0.0,
"total": 1.710100250531652e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 1169522.0788244556
}
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 955.994112,
"max_global_vram": null,
"max_process_vram": null,
"max_reserved": null,
"max_allocated": null
},
"latency": {
"unit": "s",
"count": 15,
"total": 0.3997016369999642,
"mean": 0.026646775799997614,
"stdev": 0.0007918985761803657,
"p50": 0.026372221999992007,
"p90": 0.027631125999999995,
"p95": 0.028357824800005458,
"p99": 0.028737838559989654,
"values": [
0.028832841999985703,
0.026538831999999957,
0.02665899599998056,
0.026846445999979096,
0.026372221999992007,
0.0263706579999905,
0.026654519000004484,
0.026315214999982572,
0.026356442000007974,
0.026503305999995064,
0.026308583999991697,
0.026334702000013976,
0.028154246000013927,
0.025596064000012575,
0.0258585630000141
]
},
"throughput": {
"unit": "tokens/s",
"value": 37.527992411002664
},
"energy": {
"unit": "kWh",
"cpu": 9.915335200451032e-07,
"ram": 4.143897075703743e-08,
"gpu": 0.0,
"total": 1.0329724908021417e-06
},
"efficiency": {
"unit": "tokens/kWh",
"value": 968079.9913882146
}
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 14,
"total": 0.9778543760000105,
"mean": 0.0698467411428579,
"stdev": 0.0009704698428335558,
"p50": 0.06972004299998957,
"p90": 0.07039668429999893,
"p95": 0.07130081135000524,
"p99": 0.07245785347000833,
"values": [
0.0727471140000091,
0.06950713099999462,
0.07009253199998966,
0.06936220000000048,
0.06916506300001402,
0.07010420399998907,
0.06986435700000015,
0.06962508099999809,
0.0693107940000175,
0.06981500499998106,
0.06948357699999974,
0.07052203300000315,
0.07009508699999856,
0.06816019800001527
]
},
"throughput": {
"unit": "tokens/s",
"value": 14.317060232698543
},
"energy": null,
"efficiency": null
}
}
}