IlyasMoutawwakil
HF staff
Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub
fdcb7b7
verified
| { | |
| "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": "347e13ca9f7f904f55669603cfb9f0b6c7e8672c", | |
| "transformers_version": "4.41.1", | |
| "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": 955.838464, | |
| "max_global_vram": null, | |
| "max_process_vram": null, | |
| "max_reserved": null, | |
| "max_allocated": null | |
| }, | |
| "latency": { | |
| "unit": "s", | |
| "count": 13, | |
| "total": 0.6638927820000333, | |
| "mean": 0.0510686755384641, | |
| "stdev": 0.002045444067699494, | |
| "p50": 0.05120662400000242, | |
| "p90": 0.05376475900000628, | |
| "p95": 0.054256579000008284, | |
| "p99": 0.05438413660001174, | |
| "values": [ | |
| 0.05207829200000447, | |
| 0.05441602600001261, | |
| 0.05415028100000541, | |
| 0.0515553809999858, | |
| 0.048570006000005606, | |
| 0.05222267100000977, | |
| 0.05146354100000394, | |
| 0.05041946099998995, | |
| 0.048943198000017674, | |
| 0.05120662400000242, | |
| 0.051157893000009835, | |
| 0.05108318399999234, | |
| 0.0466262239999935 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 39.16295026084302 | |
| }, | |
| "energy": { | |
| "unit": "kWh", | |
| "cpu": 1.766204891573404e-06, | |
| "ram": 7.381309496177772e-08, | |
| "gpu": 0.0, | |
| "total": 1.8400179865351817e-06 | |
| }, | |
| "efficiency": { | |
| "unit": "tokens/kWh", | |
| "value": 1086945.8965268433 | |
| } | |
| }, | |
| "decode": { | |
| "memory": { | |
| "unit": "MB", | |
| "max_ram": 955.838464, | |
| "max_global_vram": null, | |
| "max_process_vram": null, | |
| "max_reserved": null, | |
| "max_allocated": null | |
| }, | |
| "latency": { | |
| "unit": "s", | |
| "count": 13, | |
| "total": 0.40678622899997663, | |
| "mean": 0.031291248384613585, | |
| "stdev": 0.001711399813907373, | |
| "p50": 0.03150826099999904, | |
| "p90": 0.03236894979999079, | |
| "p95": 0.03312458519999381, | |
| "p99": 0.033997589040000095, | |
| "values": [ | |
| 0.032397081999988586, | |
| 0.03421584000000166, | |
| 0.03225642099999959, | |
| 0.03081424100000163, | |
| 0.0313733799999909, | |
| 0.032005884999989576, | |
| 0.032052802000009706, | |
| 0.03150826099999904, | |
| 0.03156042799997749, | |
| 0.031312657000000854, | |
| 0.031474787999997034, | |
| 0.026685342000007495, | |
| 0.029129102000013063 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 31.95781733309548 | |
| }, | |
| "energy": { | |
| "unit": "kWh", | |
| "cpu": 1.089805969293567e-06, | |
| "ram": 4.554324453157282e-08, | |
| "gpu": 0.0, | |
| "total": 1.1353492138251402e-06 | |
| }, | |
| "efficiency": { | |
| "unit": "tokens/kWh", | |
| "value": 880786.2707112546 | |
| } | |
| }, | |
| "per_token": { | |
| "memory": null, | |
| "latency": { | |
| "unit": "s", | |
| "count": 12, | |
| "total": 0.986620572000021, | |
| "mean": 0.08221838100000174, | |
| "stdev": 0.0033235099788530573, | |
| "p50": 0.08248187099998461, | |
| "p90": 0.08624422410000818, | |
| "p95": 0.08742922315001067, | |
| "p99": 0.08836903983001036, | |
| "values": [ | |
| 0.08860399400001029, | |
| 0.08646804700001098, | |
| 0.08247610999998756, | |
| 0.07991193800000929, | |
| 0.08422981799998297, | |
| 0.08354884300001686, | |
| 0.08192640999999412, | |
| 0.08056282600000486, | |
| 0.08248763199998166, | |
| 0.08264841100000808, | |
| 0.07797228399999767, | |
| 0.07578425900001662 | |
| ] | |
| }, | |
| "throughput": { | |
| "unit": "tokens/s", | |
| "value": 12.16273037534002 | |
| }, | |
| "energy": null, | |
| "efficiency": null | |
| } | |
| } | |
| } |