IlyasMoutawwakil's picture
Upload cuda_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
00c3eef verified
raw
history blame
13.5 kB
{
"config": {
"name": "cuda_inference_transformers_text-generation_openai-community/gpt2",
"backend": {
"name": "pytorch",
"version": "2.2.2+rocm5.7",
"_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": false,
"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",
"start_method": "spawn"
},
"environment": {
"cpu": " AMD EPYC 7763 64-Core Processor",
"cpu_count": 128,
"cpu_ram_mb": 1082015.236096,
"system": "Linux",
"machine": "x86_64",
"platform": "Linux-5.15.0-84-generic-x86_64-with-glibc2.35",
"processor": "x86_64",
"python_version": "3.10.12",
"gpu": [
"Advanced Micro Devices, Inc. [AMD/ATI]"
],
"gpu_count": 1,
"gpu_vram_mb": 68702699520,
"optimum_benchmark_version": "0.2.0",
"optimum_benchmark_commit": "6fd377459e287bb09e9383ba2516b1b2a271a562",
"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": "0.9.16",
"timm_commit": null,
"peft_version": null,
"peft_commit": null
}
},
"report": {
"prefill": {
"memory": {
"unit": "MB",
"max_ram": 1047.535616,
"max_global_vram": 1118.416896,
"max_process_vram": 265612.439552,
"max_reserved": 725.614592,
"max_allocated": 512.832512
},
"latency": {
"unit": "s",
"count": 58,
"total": 0.501284252166748,
"mean": 0.00864283193390945,
"stdev": 0.0006630042210832898,
"p50": 0.00841030502319336,
"p90": 0.009131984901428223,
"p95": 0.009151904249191283,
"p99": 0.010536824684143065,
"values": [
0.012341577529907227,
0.008249106407165528,
0.008205586433410645,
0.008510065078735352,
0.008168306350708008,
0.00813870620727539,
0.00814478588104248,
0.008347504615783692,
0.008314226150512695,
0.008157585144042969,
0.008179025650024414,
0.008336146354675293,
0.008338226318359374,
0.008473105430603027,
0.008083986282348634,
0.008072465896606444,
0.008089105606079101,
0.00825390625,
0.008062545776367187,
0.00826958465576172,
0.008206066131591796,
0.008076465606689454,
0.008078226089477539,
0.008128786087036132,
0.008085745811462403,
0.008067505836486816,
0.008081746101379395,
0.008112626075744629,
0.00806558609008789,
0.00809134578704834,
0.008074706077575683,
0.008296785354614257,
0.009116144180297851,
0.008836785316467284,
0.009108624458312987,
0.009090383529663086,
0.009123663902282714,
0.009126705169677734,
0.009103504180908203,
0.009106863975524902,
0.009076784133911133,
0.009124943733215332,
0.009096464157104493,
0.009117584228515625,
0.009083024978637695,
0.009175344467163086,
0.009098223686218262,
0.009106063842773437,
0.009147824287414551,
0.009149744033813476,
0.009144304275512695,
0.009124943733215332,
0.009107024192810058,
0.009164145469665527,
0.00878990364074707,
0.008793265342712402,
0.00851662540435791,
0.008754224777221679
]
},
"throughput": {
"unit": "tokens/s",
"value": 231.40563362723302
},
"energy": null,
"efficiency": null
},
"decode": {
"memory": {
"unit": "MB",
"max_ram": 1047.535616,
"max_global_vram": 1118.408704,
"max_process_vram": 281294.352384,
"max_reserved": 725.614592,
"max_allocated": 513.056256
},
"latency": {
"unit": "s",
"count": 58,
"total": 0.502205677986145,
"mean": 0.008658718585968018,
"stdev": 0.0004390677993088848,
"p50": 0.008537344932556153,
"p90": 0.00916436767578125,
"p95": 0.009191783714294435,
"p99": 0.009239040069580078,
"values": [
0.008550225257873535,
0.008258384704589844,
0.008178385734558105,
0.00839118480682373,
0.008186705589294433,
0.00817614459991455,
0.008775505065917968,
0.008520625114440917,
0.00817486572265625,
0.008344945907592773,
0.008227825164794922,
0.008235344886779785,
0.008263344764709473,
0.008161106109619141,
0.008136305809020996,
0.00826958465576172,
0.008114066123962402,
0.008377264976501465,
0.008123826026916504,
0.008329265594482422,
0.008115345001220704,
0.00812398624420166,
0.008208786010742188,
0.008433904647827149,
0.008128625869750977,
0.008107665061950684,
0.008900625228881836,
0.008307825088500977,
0.008337586402893067,
0.00814478588104248,
0.008128786087036132,
0.009188783645629884,
0.009213583946228027,
0.009156304359436035,
0.009141743659973145,
0.009166383743286132,
0.009109265327453613,
0.009139663696289062,
0.00914078426361084,
0.009110063552856444,
0.009272784233093261,
0.009163503646850585,
0.00914254379272461,
0.009139823913574218,
0.009128144264221191,
0.009150223731994629,
0.009170063972473145,
0.009208784103393555,
0.009140144348144532,
0.00909102439880371,
0.009136783599853515,
0.009139823913574218,
0.009120784759521484,
0.008986543655395508,
0.008934544563293458,
0.008853424072265624,
0.008802864074707031,
0.00852446460723877
]
},
"throughput": {
"unit": "tokens/s",
"value": 115.49053016003558
},
"energy": null,
"efficiency": null
},
"per_token": {
"memory": null,
"latency": {
"unit": "s",
"count": 57,
"total": 0.9844515552520751,
"mean": 0.017271079916703073,
"stdev": 0.0008708116033835836,
"p50": 0.016969249725341796,
"p90": 0.018308320236206054,
"p95": 0.018355776596069337,
"p99": 0.018384454040527343,
"values": [
0.01654429054260254,
0.016426851272583008,
0.01693404960632324,
0.016387811660766603,
0.016362850189208984,
0.01688717269897461,
0.016969249725341796,
0.016526531219482422,
0.016527332305908204,
0.016434051513671874,
0.01661293029785156,
0.016646371841430665,
0.01666813087463379,
0.01623549270629883,
0.016393571853637694,
0.01622669219970703,
0.016660451889038086,
0.01622093200683594,
0.016595651626586914,
0.0163847713470459,
0.01622669219970703,
0.016274852752685545,
0.016635011672973633,
0.016248931884765627,
0.01619997215270996,
0.01701388931274414,
0.01645197105407715,
0.01641853141784668,
0.016273252487182616,
0.016240291595458985,
0.017449247360229493,
0.01837628746032715,
0.018030847549438477,
0.018278047561645507,
0.01829452705383301,
0.018260128021240236,
0.018302528381347657,
0.01828220748901367,
0.018242368698120116,
0.018394847869873047,
0.01831884765625,
0.018272127151489258,
0.018294687271118163,
0.018232128143310548,
0.018360128402709962,
0.01829212760925293,
0.01835468864440918,
0.01829372787475586,
0.018306848526000976,
0.01831052780151367,
0.018298847198486328,
0.018263647079467774,
0.018207647323608397,
0.017730047225952148,
0.017725568771362304,
0.01730828857421875,
0.01734204864501953
]
},
"throughput": {
"unit": "tokens/s",
"value": 57.90025897760381
},
"energy": null,
"efficiency": null
}
}
}