{
    "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",
            "numactl": false,
            "numactl_kwargs": {},
            "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": "a426049b3dc97f6bbcbc7a79743065432e963ac0",
            "transformers_version": "4.41.0",
            "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": 955.809792,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 14,
                "total": 0.6562045120000448,
                "mean": 0.04687175085714606,
                "stdev": 0.0023833333770636347,
                "p50": 0.047949733500018965,
                "p90": 0.04874958939999487,
                "p95": 0.04946343594998268,
                "p99": 0.0504996799899817,
                "values": [
                    0.04812937999997757,
                    0.04783337499998197,
                    0.04876596399998334,
                    0.04822122300004139,
                    0.048711382000021786,
                    0.047962227000027724,
                    0.048238454999989244,
                    0.047937240000010206,
                    0.050758740999981455,
                    0.04314865099996723,
                    0.04320428599999104,
                    0.04311434700002792,
                    0.04584505700000818,
                    0.04433418400003575
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 42.669624313704944
            },
            "energy": {
                "unit": "kWh",
                "cpu": 1.6877410588441071e-06,
                "ram": 7.053051078093138e-08,
                "gpu": 0.0,
                "total": 1.7582715696250386e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 1137480.7137594288
            }
        },
        "decode": {
            "memory": {
                "unit": "MB",
                "max_ram": 955.809792,
                "max_global_vram": null,
                "max_process_vram": null,
                "max_reserved": null,
                "max_allocated": null
            },
            "latency": {
                "unit": "s",
                "count": 14,
                "total": 0.40132704699988153,
                "mean": 0.02866621764284868,
                "stdev": 0.0018734309703798524,
                "p50": 0.029927428499973985,
                "p90": 0.030496800399970426,
                "p95": 0.03056522004997646,
                "p99": 0.030585470410002244,
                "values": [
                    0.03059053300000869,
                    0.030368957999996837,
                    0.030183729000043513,
                    0.030551589999959106,
                    0.03033209799997394,
                    0.03006684999996878,
                    0.02978800699997919,
                    0.030336245999990297,
                    0.026508308000018133,
                    0.02643478900000673,
                    0.026178469000001314,
                    0.026568991999965874,
                    0.0267787539999631,
                    0.026639724000006026
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 34.884267344194555
            },
            "energy": {
                "unit": "kWh",
                "cpu": 9.998117994379114e-07,
                "ram": 4.178808104595079e-08,
                "gpu": 0.0,
                "total": 1.0415998804838628e-06
            },
            "efficiency": {
                "unit": "tokens/kWh",
                "value": 960061.5540925964
            }
        },
        "per_token": {
            "memory": null,
            "latency": {
                "unit": "s",
                "count": 13,
                "total": 0.9791904059999865,
                "mean": 0.07532233892307588,
                "stdev": 0.003934994274308058,
                "p50": 0.07799288999996179,
                "p90": 0.07891431559999092,
                "p95": 0.07899124179997444,
                "p99": 0.07904548755996757,
                "values": [
                    0.07822192899999436,
                    0.07894603699998015,
                    0.07878743000003396,
                    0.07905904899996585,
                    0.07807059500004243,
                    0.07799288999996179,
                    0.07831641699999636,
                    0.07746394800000189,
                    0.06959744700003512,
                    0.0693835949999766,
                    0.06968472099998735,
                    0.07266250399999308,
                    0.07100384400001758
                ]
            },
            "throughput": {
                "unit": "tokens/s",
                "value": 13.276273869047875
            },
            "energy": null,
            "efficiency": null
        }
    }
}