{ "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": "0.2.0", "optimum_benchmark_commit": "b04fb3c909a5873eadf03d7b46ccfac63afcdf9e", "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": 957.083648, "max_global_vram": null, "max_process_vram": null, "max_reserved": null, "max_allocated": null }, "latency": { "unit": "s", "count": 13, "total": 0.6171481320000396, "mean": 0.04747293323077228, "stdev": 0.0019100159827317189, "p50": 0.04810255399999619, "p90": 0.048698938000001135, "p95": 0.04880850580000242, "p99": 0.048909791559999576, "values": [ 0.048724101000004794, 0.04856443299999569, 0.048935112999998864, 0.04810255399999619, 0.047646637000013925, 0.048101783000021214, 0.048229791000011346, 0.04844186500000092, 0.0485982859999865, 0.047852468999991515, 0.047792137000016055, 0.042878946999991285, 0.0432800160000113 ] }, "throughput": { "unit": "tokens/s", "value": 42.129269541398095 }, "energy": { "unit": "kWh", "cpu": 1.7578606444280503e-06, "ram": 7.346402652153162e-08, "gpu": 0.0, "total": 1.8313246709495818e-06 }, "efficiency": { "unit": "tokens/kWh", "value": 1092105.6390088145 } }, "decode": { "memory": { "unit": "MB", "max_ram": 957.083648, "max_global_vram": null, "max_process_vram": null, "max_reserved": null, "max_allocated": null }, "latency": { "unit": "s", "count": 13, "total": 0.38532145100001003, "mean": 0.029640111615385385, "stdev": 0.0015671268901592899, "p50": 0.030042218000005505, "p90": 0.030686499999995932, "p95": 0.03133926799999926, "p99": 0.03201551120000431, "values": [ 0.03077573199999506, 0.030107358999998723, 0.030042218000005505, 0.029908879999993587, 0.029936149999997497, 0.030166428999990558, 0.030118679999986853, 0.032184572000005573, 0.030329571999999416, 0.02970274700001596, 0.029489949999998544, 0.026399023000010402, 0.02616013900001235 ] }, "throughput": { "unit": "tokens/s", "value": 33.73806458545611 }, "energy": { "unit": "kWh", "cpu": 9.855457396514947e-07, "ram": 4.118853906375675e-08, "gpu": 0.0, "total": 1.026734278715252e-06 }, "efficiency": { "unit": "tokens/kWh", "value": 973961.8329012016 } }, "per_token": { "memory": null, "latency": { "unit": "s", "count": 12, "total": 0.9233822879999991, "mean": 0.07694852399999992, "stdev": 0.003457137844751448, "p50": 0.07813803450000023, "p90": 0.07897906219999129, "p95": 0.07973946410000395, "p99": 0.0804746504200159, "values": [ 0.07872747600001162, 0.07898756899999171, 0.0780329239999844, 0.07760936700000798, 0.07824314500001606, 0.0783940259999838, 0.0806584470000189, 0.07890250099998752, 0.07759917699999619, 0.07733504699999116, 0.06943643500000007, 0.0694561740000097 ] }, "throughput": { "unit": "tokens/s", "value": 12.995700866205063 }, "energy": null, "efficiency": null } } }