Hugging Face Downloader

A GUI tool for downloading models, datasets, and spaces from Hugging Face.

中文文档

Features

• Download entire repositories or single files
• Support for models, datasets, and spaces
• Multi-language support (English, Chinese)
• Custom save path
• Progress tracking
• Private repository support with token
• Automatic retry and fallback mechanisms
• SSL verification bypass for regions with connection issues

Installation

1. Clone this repository:

git clone https://github.com/yourusername/hf_downloader.git
cd hf_downloader

2. Install dependencies:

pip install -r requirements.txt

Usage

1. Run the application:

python gui.py

2. Enter the repository information:

   • Repository ID (e.g., bert-base-chinese or username/repo-name)
   • Select repository type (model, dataset, or space)
   • Choose between downloading entire repository or single file
   • For single file download, specify filename and optional subfolder
   • Optionally provide HF token for private repositories
   • Choose custom save path or use default cache directory

3. Click "Start Download" and wait for completion

Requirements

• Python 3.6+
• PyQt6
• huggingface_hub
• requests
• python-dotenv
• dnspython

Known Issues

• SSL certificate verification might fail in some regions
• Download might be slow in certain locations
• Some private repositories might require additional authentication

Contributing

Feel free to open issues or submit pull requests for any bugs or improvements.