{
    "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": 16757.3504,
            "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": "1ba65d84c92c93ea19ab2d2293c92bced987bae2",
            "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": 955.920384,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.6252996339999868,
                "mean": 0.04809997184615283,
                "stdev": 0.0027160685920652595,
                "p50": 0.04880595700001322,
                "p90": 0.050437242199996034,
                "p95": 0.051055869599986176,
                "p99": 0.051753131519972156,
                "values": [
                    0.04907889400004706,
                    0.05047481799999787,
                    0.05192744699996865,
                    0.048919436999995014,
                    0.04862169399996219,
                    0.050286938999988706,
                    0.04951671700001725,
                    0.048537347000035425,
                    0.04880595700001322,
                    0.04817735899996478,
                    0.04577221599998893,
                    0.04237234400000034,
                    0.042808465000007345
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 41.58006591764701
            },
            "energy": {
                "unit": "kWh",
                "cpu": 1.6686885997101114e-06,
                "ram": 6.97371097092435e-08,
                "gpu": 0.0,
                "total": 1.738425709419355e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 1150466.188554018
            }
        },
        "decode": {
            "memory": {
                "unit": "MB",
                "max_ram": 956.051456,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.3867308330000583,
                "mean": 0.0297485256153891,
                "stdev": 0.0018391550381465776,
                "p50": 0.030680057000040506,
                "p90": 0.030932181000036963,
                "p95": 0.03132587080003759,
                "p99": 0.03178866376003498,
                "values": [
                    0.03071368899998106,
                    0.031904362000034325,
                    0.03094021000003977,
                    0.029902933000016674,
                    0.030900065000025734,
                    0.030680057000040506,
                    0.030879346999995505,
                    0.03086412899995139,
                    0.030143679999980577,
                    0.029886853999983032,
                    0.02628012600001739,
                    0.025624485999969693,
                    0.02801089500002263
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 33.61511131438036
            },
            "energy": {
                "unit": "kWh",
                "cpu": 1.0910912244408226e-06,
                "ram": 4.55980283176653e-08,
                "gpu": 0.0,
                "total": 1.1366892527584873e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 879747.9148969047
            }
        },
        "per_token": {
            "memory": null,
            "latency": {
                "unit": "s",
                "count": 12,
                "total": 0.932674392000024,
                "mean": 0.077722866000002,
                "stdev": 0.004538825700259946,
                "p50": 0.07918224900001292,
                "p90": 0.08226358920001076,
                "p95": 0.08262397670001746,
                "p99": 0.08283854094002209,
                "values": [
                    0.08240453600001274,
                    0.08289218200002324,
                    0.07886698300001171,
                    0.07952760099999523,
                    0.0809950679999929,
                    0.08042620999998462,
                    0.07941543200001888,
                    0.07894906600000695,
                    0.07807533199996897,
                    0.07229349099998217,
                    0.0680083800000375,
                    0.07082011099998908
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 12.866226523349953
            },
            "energy": null,
            "efficiency": null
        }
    }
}