Package profile
optimum
- Summary: Optimum Library is an extension of the Hugging Face Transformers library, providing a framework to integrate third-party libraries from Hardware Partners and interface with their specific functionality.
- Author: HuggingFace Inc. Special Ops Team
- Homepage: https://github.com/huggingface/optimum
- Source: https://github.com/huggingface/optimum (Repo profile)
- Number of releases: 79
- First release: 0.0.1 on 2021-09-14
- Latest release: 1.26.1 on 2025-06-13
Releases
PyPI Downloads
Dependencies
Optimum has 44 dependencies, 39 of which optional.| Dependency | Optional |
|---|---|
| huggingface-hub | false |
| numpy | false |
| packaging | false |
| torch | false |
| transformers | false |
| accelerate | true |
| black | true |
| datasets | true |
| einops | true |
| evaluate | true |
| h5py | true |
| hf-xet | true |
| onnx | true |
| onnxruntime | true |
| onnxruntime-gpu | true |
| onnxruntime-training | true |
| onnxslim | true |
| optimum-amd | true |
| optimum-furiosa | true |
| optimum-graphcore | true |
| optimum-habana | true |
| optimum-intel | true |
| optimum-neuron | true |
| optimum-quanto | true |
| optuna | true |
| parameterized | true |
| Pillow | true |
| protobuf | true |
| pytest | true |
| pytest-xdist | true |
| requests | true |
| rjieba | true |
| ruff | true |
| sacremoses | true |
| scikit-learn | true |
| sentencepiece | true |
| seqeval | true |
| tensorflow | true |
| tf2onnx | true |
| timm | true |
| torch-ort | true |
| torchaudio | true |
| torchvision | true |
| tqdm | true |