{
    "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": "0b24af9d7b7751f74b160dfade73ef78e10964d6",
            "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": 945.07008,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.6660168779999651,
                "mean": 0.05123206753845885,
                "stdev": 0.0013537270955456635,
                "p50": 0.051345710999981975,
                "p90": 0.052579589600003376,
                "p95": 0.053123269000002436,
                "p99": 0.05369348500000342,
                "values": [
                    0.05264808900000162,
                    0.053836039000003666,
                    0.05033808999999678,
                    0.0520746570000199,
                    0.05156406399999014,
                    0.04888016900000025,
                    0.051345710999981975,
                    0.052204220999982454,
                    0.050188006000013274,
                    0.04962437299997191,
                    0.05122551499999872,
                    0.049782351999994034,
                    0.05230559200001039
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 39.03804972342062
            },
            "energy": {
                "unit": "kWh",
                "cpu": 1.7559961996216708e-06,
                "ram": 7.338199189919322e-08,
                "gpu": 0.0,
                "total": 1.829378191520864e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 1093267.6519650037
            }
        },
        "decode": {
            "memory": {
                "unit": "MB",
                "max_ram": 945.07008,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.4163931050000258,
                "mean": 0.03203023884615583,
                "stdev": 0.0005299569319006806,
                "p50": 0.0321998800000074,
                "p90": 0.03258501220000767,
                "p95": 0.03261922660000778,
                "p99": 0.03264701811999998,
                "values": [
                    0.032391270999994504,
                    0.032653965999998036,
                    0.032371984999997494,
                    0.03254079299998125,
                    0.031109482000005073,
                    0.031729321000000255,
                    0.03259606700001427,
                    0.03239731200000051,
                    0.03109188999999901,
                    0.03193028100000106,
                    0.03136576600002172,
                    0.032015091000005214,
                    0.0321998800000074
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 31.220497755358355
            },
            "energy": {
                "unit": "kWh",
                "cpu": 9.796366791410338e-07,
                "ram": 4.09472447039755e-08,
                "gpu": 0.0,
                "total": 1.020583923845009e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 979831.2286093436
            }
        },
        "per_token": {
            "memory": null,
            "latency": {
                "unit": "s",
                "count": 12,
                "total": 0.997634010000013,
                "mean": 0.08313616750000108,
                "stdev": 0.001688979378226582,
                "p50": 0.08270934799999452,
                "p90": 0.08467333029999793,
                "p95": 0.08553187815000171,
                "p99": 0.08637099643000454,
                "values": [
                    0.08658077600000524,
                    0.0826527970000086,
                    0.08467368899999883,
                    0.08272227800000564,
                    0.08059725799998319,
                    0.08393735000001357,
                    0.08467010199998981,
                    0.08133652300000449,
                    0.08145419600000992,
                    0.0826964179999834,
                    0.08176083400002199,
                    0.08455178899998828
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 12.028459214216088
            },
            "energy": null,
            "efficiency": null
        }
    }
}