{
    "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": "b92e3e6e00670fa1351d9aef1be896254e5f33e0",
            "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": 956.776448,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.6306061849999764,
                "mean": 0.048508168076921265,
                "stdev": 0.0025299312995963946,
                "p50": 0.04884615000000281,
                "p90": 0.05117744780000635,
                "p95": 0.051456715000000486,
                "p99": 0.051469338999987714,
                "values": [
                    0.05147249499998452,
                    0.05144619500001113,
                    0.05010245899998722,
                    0.04942166400002179,
                    0.048712805000008075,
                    0.04883560499999362,
                    0.04879463899999337,
                    0.04958431800000085,
                    0.04887833499998351,
                    0.04884615000000281,
                    0.04850446699998656,
                    0.04329423600000837,
                    0.04271281699999463
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 41.23016966603487
            },
            "energy": {
                "unit": "kWh",
                "cpu": 1.7326936629659314e-06,
                "ram": 7.241119269548088e-08,
                "gpu": 0.0,
                "total": 1.8051048556614125e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 1107968.8771138869
            }
        },
        "decode": {
            "memory": {
                "unit": "MB",
                "max_ram": 956.776448,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.39762465899997324,
                "mean": 0.030586512230767172,
                "stdev": 0.0017609820449105583,
                "p50": 0.030928404999997383,
                "p90": 0.03238657140000214,
                "p95": 0.03276847440000097,
                "p99": 0.03293954448000136,
                "values": [
                    0.03142919300000813,
                    0.03262591600000064,
                    0.03298231200000146,
                    0.031065571999988606,
                    0.030479193999980225,
                    0.03103959399999212,
                    0.030451370999998062,
                    0.030726447000006374,
                    0.030928404999997383,
                    0.030871754000003193,
                    0.03136871099999894,
                    0.02654615199998034,
                    0.027110038000017767
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 32.694149383730434
            },
            "energy": {
                "unit": "kWh",
                "cpu": 9.984009622951638e-07,
                "ram": 4.1725385001695355e-08,
                "gpu": 0.0,
                "total": 1.040126347296859e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 961421.660550046
            }
        },
        "per_token": {
            "memory": null,
            "latency": {
                "unit": "s",
                "count": 12,
                "total": 0.945738963999986,
                "mean": 0.07881158033333217,
                "stdev": 0.004246475645647794,
                "p50": 0.07981963950000193,
                "p90": 0.08278326559999982,
                "p95": 0.08350750784999547,
                "p99": 0.08397796156999278,
                "values": [
                    0.0840955749999921,
                    0.08302636199999824,
                    0.08059539800001403,
                    0.07922230599999125,
                    0.07987344500000404,
                    0.07926495600000294,
                    0.0803406109999969,
                    0.07976519299998586,
                    0.07976583399999981,
                    0.08007023700000104,
                    0.06988299700000766,
                    0.06983604999999216
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 12.688490647827614
            },
            "energy": null,
            "efficiency": null
        }
    }
}