{ "config": { "name": "cuda_inference_transformers_text-generation_openai-community/gpt2", "backend": { "name": "pytorch", "version": "2.3.0+cu121", "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend", "task": "text-generation", "library": "transformers", "model": "openai-community/gpt2", "processor": "openai-community/gpt2", "device": "cuda", "device_ids": "0", "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": true, "device_isolation_action": "error", "numactl": false, "numactl_kwargs": {}, "start_method": "spawn" }, "environment": { "cpu": " AMD EPYC 7R32", "cpu_count": 16, "cpu_ram_mb": 66697.29792, "system": "Linux", "machine": "x86_64", "platform": "Linux-5.10.215-203.850.amzn2.x86_64-x86_64-with-glibc2.35", "processor": "x86_64", "python_version": "3.10.12", "gpu": [ "NVIDIA A10G" ], "gpu_count": 1, "gpu_vram_mb": 24146608128, "optimum_benchmark_version": "0.2.1", "optimum_benchmark_commit": null, "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": 1059.663872, "max_global_vram": 1365.77024, "max_process_vram": 0.0, "max_reserved": 725.614592, "max_allocated": 522.0096 }, "latency": { "unit": "s", "count": 56, "total": 0.49548483180999775, "mean": 0.008847943425178527, "stdev": 0.0003823576996123477, "p50": 0.00883678388595581, "p90": 0.009244911670684815, "p95": 0.009345799684524536, "p99": 0.009714172887802125, "values": [ 0.009844096183776855, 0.00950767993927002, 0.009291839599609376, 0.009050368309020996, 0.008842432022094726, 0.00917078399658203, 0.00923961639404297, 0.008888416290283203, 0.009242719650268554, 0.009139519691467285, 0.009239744186401368, 0.00910643196105957, 0.008887328147888184, 0.009030112266540527, 0.009247103691101074, 0.009006112098693848, 0.009148768424987793, 0.008853823661804198, 0.008742560386657714, 0.009009471893310547, 0.00906668758392334, 0.009281472206115722, 0.009145119667053222, 0.00908790397644043, 0.008753984451293946, 0.0091527681350708, 0.008461503982543946, 0.008831135749816895, 0.009116543769836425, 0.009223648071289062, 0.009607872009277344, 0.008920767784118652, 0.00879638385772705, 0.008318655967712403, 0.008311488151550294, 0.00829196834564209, 0.008719840049743653, 0.00880031967163086, 0.008745023727416992, 0.008605504035949707, 0.008595647811889649, 0.00857481575012207, 0.00819548797607422, 0.008157440185546875, 0.008244735717773437, 0.008170047760009765, 0.00816368007659912, 0.008209024429321289, 0.008789471626281738, 0.008805312156677247, 0.00872976016998291, 0.00856009578704834, 0.008559743881225585, 0.00881721591949463, 0.008588224411010743, 0.00859660816192627 ] }, "throughput": { "unit": "tokens/s", "value": 226.04122832754726 }, "energy": { "unit": "kWh", "cpu": 1.0232712602478336e-07, "ram": 5.592363140435893e-08, "gpu": 1.8009353487930782e-07, "total": 3.383442923084501e-07 }, "efficiency": { "unit": "tokens/kWh", "value": 5911138.581219832 } }, "decode": { "memory": { "unit": "MB", "max_ram": 1059.663872, "max_global_vram": 1365.77024, "max_process_vram": 0.0, "max_reserved": 725.614592, "max_allocated": 522.46016 }, "latency": { "unit": "s", "count": 56, "total": 0.5067465600967407, "mean": 0.009049045716013226, "stdev": 0.00036576571201634524, "p50": 0.009077712059020995, "p90": 0.00945030403137207, "p95": 0.00950822377204895, "p99": 0.009709664058685303, "values": [ 0.00972761631011963, 0.009694975852966308, 0.009514464378356933, 0.009133440017700195, 0.009223808288574218, 0.00950614356994629, 0.009325152397155761, 0.009372127532958985, 0.009502016067504882, 0.009438400268554687, 0.009308320045471191, 0.009398079872131348, 0.009117152214050293, 0.009415488243103027, 0.00931385612487793, 0.009256799697875977, 0.009357407569885253, 0.009164928436279297, 0.00898793601989746, 0.009384639739990234, 0.009239168167114257, 0.009462207794189452, 0.009377599716186524, 0.009333439826965332, 0.009251551628112793, 0.009008607864379882, 0.008610367774963379, 0.009296223640441895, 0.009381471633911133, 0.009048416137695312, 0.009289119720458984, 0.009038528442382813, 0.008952256202697753, 0.008461248397827149, 0.008433247566223144, 0.008663455963134765, 0.00892959976196289, 0.009107007980346679, 0.008802207946777343, 0.008841983795166016, 0.008807807922363282, 0.008838815689086914, 0.008339648246765136, 0.008372096061706543, 0.008405376434326171, 0.008353728294372559, 0.008347200393676759, 0.008406975746154784, 0.008979424476623536, 0.00894105625152588, 0.00891654396057129, 0.00888047981262207, 0.008896991729736329, 0.008998656272888183, 0.008787424087524413, 0.008803872108459473 ] }, "throughput": { "unit": "tokens/s", "value": 110.50889026125662 }, "energy": { "unit": "kWh", "cpu": 9.2995267369124e-08, "ram": 5.087002149400585e-08, "gpu": 1.6844499531741864e-07, "total": 3.123102841805484e-07 }, "efficiency": { "unit": "tokens/kWh", "value": 3201943.8700964907 } }, "per_token": { "memory": null, "latency": { "unit": "s", "count": 55, "total": 0.9844951095581052, "mean": 0.017899911082874642, "stdev": 0.0007090343583201165, "p50": 0.01804083251953125, "p90": 0.018659328079223634, "p95": 0.018819276237487793, "p99": 0.019072307014465332, "values": [ 0.01924095916748047, 0.0188723201751709, 0.01822412872314453, 0.01810534477233887, 0.018679807662963867, 0.01862553596496582, 0.018284543991088868, 0.018767871856689454, 0.01862860870361328, 0.01860403251647949, 0.018521087646484375, 0.01804083251953125, 0.018457599639892578, 0.01861734390258789, 0.01829478454589844, 0.01853446388244629, 0.018067392349243164, 0.017772544860839845, 0.01841971206665039, 0.018321407318115233, 0.01879654312133789, 0.018561023712158203, 0.018449407577514648, 0.018053119659423827, 0.018269216537475586, 0.017101791381835936, 0.018089984893798827, 0.018509855270385744, 0.018392032623291015, 0.018928640365600585, 0.01801625633239746, 0.017772544860839845, 0.01682329559326172, 0.016764928817749023, 0.016978944778442383, 0.017674240112304687, 0.017936384201049805, 0.01757900810241699, 0.017481727600097655, 0.017427455902099608, 0.01744076728820801, 0.016563199996948243, 0.01656524848937988, 0.016647167205810547, 0.016574464797973632, 0.016538623809814454, 0.01662668800354004, 0.01779302406311035, 0.017777664184570312, 0.01764761543273926, 0.01744588851928711, 0.01752275276184082, 0.017828800201416015, 0.01742233657836914, 0.01741414451599121 ] }, "throughput": { "unit": "tokens/s", "value": 55.86619929954449 }, "energy": null, "efficiency": null } } }