Accelerate
Hugging Face library to easily train and run models on multiple GPUs and TPUs.
ML and data science PyPI Apache-2.0
How Accelerate is used
Accelerate is a dependency in 16 of the 6,208 public GitHub projects set up for AI coding agents that we scanned, too few to describe it from them alone. Our scan also samples projects by dependency (Solana 2,661, Ethereum and EVM 1,496, 3D and games 2,818, agent frameworks and SDKs 10,983); counting those, it appears in 284 projects, and what it is used with below comes from all of them. Projects that use it most often also use Hugging Face Transformers (82.7%), NumPy (80.6%), pandas (74.3%).
Accelerate popularity over time
Gid Score 65: downloads 78, use in agent projects 66, GitHub stars 50, downloads trend 57 (each 0-100). How it is calculated.
Often used with Accelerate
| Hugging Face Transformers | 82.7% of Accelerate projects · 66.7× more than average |
|---|---|
| NumPy | 80.6% of Accelerate projects · 13.8× more than average |
| pandas | 74.3% of Accelerate projects · 17.3× more than average |
| PyTorch | 74.3% of Accelerate projects · 49.1× more than average |
| Hugging Face Hub | 69.0% of Accelerate projects · 75.2× more than average |
| OpenAI Python Library | 65.5% of Accelerate projects · 16.6× more than average |
| Pillow | 62.3% of Accelerate projects · 16.3× more than average |
| Requests | 62.0% of Accelerate projects · 9.1× more than average |
From all 284 projects in our scan that use Accelerate, including the dependency samples.
By coding agent
| Claude Code | 0.3% of projects set up for it |
|---|---|
| GitHub Copilot | 0.3% of projects set up for it |
Share of public repos configured for each agent that list Accelerate as a dependency.
Install
| Package | pip install accelerate |
|---|---|
| Main stacks | Python (283), React Native / Expo (1) |
| Example projects | Paper2Poster/Paper2Poster, apconw/Aix-DB, FlagOpen/RoboBrain2.5, GAIR-NLP/DeepResearcher, OTA-Tech-AI/web-agent-protocol |
Alternatives to Accelerate
In our scan, none of the most used tools in ML and data science does the same job as Accelerate: they are add-ons, companions or tools for a different task.
More in ML and data science: NumPy pandas scikit-learn SciPy PyTorch Sentence Transformers Hugging Face Transformers NetworkX