| --- |
| license: cc0-1.0 |
| language: |
| - en |
| tags: |
| - tutorial |
| - help |
| --- |
| # Uploading Large Files to Hugging Face Hub |
|
|
| ## Prerequisites |
| - [Git](https://git-scm.com/) installed |
| - [Git LFS](https://git-lfs.com/) installed |
| - Python with `huggingface_hub` library |
| - Hugging Face account |
|
|
| ## Step-by-Step Instructions |
|
|
| ### 1. Install Required Tools |
| ```bash |
| # Install Hugging Face CLI |
| python -m pip install huggingface_hub |
| |
| # Login to Hugging Face |
| huggingface-cli login |
| ``` |
|
|
| ### 2. Create Repository |
| 1. Go to [https://huggingface.co/new](https://huggingface.co/new) |
| 2. Create a new model repository |
| 3. Copy the repository URL |
|
|
| ### 3. Clone Repository Locally |
| ```bash |
| git clone https://huggingface.co/<your-username>/<your-model-name> |
| cd <your-model-name> |
| ``` |
|
|
| ### 4. Prepare for Large File Uploads |
| ```bash |
| # Initialize Git LFS |
| git lfs install |
| |
| # Enable large file support |
| huggingface-cli lfs-enable-largefiles . |
| |
| # Track safetensor files and tokenizer.json (because they're large) |
| git lfs track "*.safetensors" "tokenizer.json" |
| ``` |
|
|
| ### 5. Add and Push Files |
|
|
| Copy all your model files to repository directory |
|
|
| Then you can run: |
| ```bash |
| git add . |
| git commit -m "Add model files" |
| git push |
| ``` |
|
|
| ## Troubleshooting |
| - Ensure files are < 5GB per file |
| - Check internet connection |
| - Verify Git LFS is correctly installed |
| - Use `git lfs status` to check file tracking |
|
|
| ## Best Practices |
| - Use descriptive commit messages |
| - Keep repository organized |
| - Include a `README.md` with model details |
