Package profile
optimum-benchmark
- Summary: Optimum-Benchmark is a unified multi-backend utility for benchmarking Transformers, Timm, Diffusers and Sentence-Transformers with full support of Optimum's hardware optimizations & quantization schemes.
- Author: HuggingFace Inc. Special Ops Team
- Homepage: https://github.com/huggingface/optimum-benchmark
- Source: https://github.com/huggingface/optimum-benchmark (Repo profile)
- Number of releases: 7
- First release: 0.2.0 on 2024-05-16
- Latest release: 0.6.0 on 2025-08-19
Releases
PyPI Downloads
Loading PyPI statistics...
Dependencies
Optimum-benchmark has 35 dependencies, 21 of which optional.| Dependency | Optional |
|---|---|
| accelerate | false |
| colorlog | false |
| datasets | false |
| flatten-dict | false |
| hf-xet | false |
| huggingface-hub | false |
| hydra-core | false |
| nvidia-ml-py | false |
| omegaconf | false |
| pandas | false |
| psutil | false |
| rich | false |
| transformers | false |
| typing-extensions | false |
| bitsandbytes | true |
| codecarbon | true |
| deepspeed | true |
| diffusers | true |
| flash-attn | true |
| gptqmodel | true |
| hydra-joblib-launcher | true |
| intel-extension-for-pytorch | true |
| llama-cpp-python | true |
| mock | true |
| optimum | true |
| peft | true |
| py-txi | true |
| pytest | true |
| ruff | true |
| sentence-transformers | true |
| timm | true |
| torch | true |
| torchao | true |
| torchcodec | true |
| vllm | true |