ComfyUI-Downloader

A powerful and user-friendly extension for ComfyUI that adds a built-in model downloader with a sleek modal interface. Download models, LoRAs, VAEs, and other assets directly from URLs without leaving ComfyUI.

Features

• 🚀 High-Speed Downloads: Uses wget on the ComfyUI server for robust large-file downloads
• 📦 Queue Management: Download multiple files with automatic queue processing
• 🎯 Smart Integration: Detects missing models from your workflow and suggests downloads
• ⚡ One-Click Missing Downloads: Queue every missing workflow model from the top-right action bar
• 🔄 Real-time Progress: Live progress tracking for queued and active downloads
• 🛡️ Security: Built-in path traversal protection and file validation
• 📂 Auto-Organization: Downloads files to the correct ComfyUI folders (checkpoints, loras, VAEs, etc.)
• ⏸️ Download Control: Cancel downloads in progress with queue management
• 🎨 Clean UI: Modal-based interface that integrates seamlessly with ComfyUI

👉 Watch Full Video

Models urls are loaded from workflow or https://github.com/Comfy-Org/ComfyUI-Manager/blob/main/model-list.json

Installation

Method 2: Manual Installation

1. Navigate to your ComfyUI custom nodes directory:

   cd ComfyUI/custom_nodes/

2. Clone this repository:

   git clone https://github.com/taigadit/ComfyUI-Downloader.git

3. Install Python dependencies:

   cd ComfyUI-Downloader
   pip install -r requirements.txt

4. Install wget if your system does not already provide it:

   # macOS (Homebrew)
   brew install wget
   
   # Ubuntu / Debian
   sudo apt-get install wget

5. Restart ComfyUI

Requirements

• Python 3.8+
• ComfyUI
• Dependencies:
• Python package:
  • aiohttp
• System dependency:
  • wget available in server PATH

Usage

Basic Usage

1. Click the "Downloader" button in the ComfyUI interface (next to the settings button)
2. The downloader modal will open
3. Use Refresh Models to scan the current workflow for referenced models
4. To download everything missing at once, click Download Missing in the top-right action bar
5. Or paste a model URL in the manual download section for a single file
6. Select the destination folder (checkpoints, loras, vae, etc.)
7. Enter a filename (supports subfolders like subfolder/model.safetensors)
8. Click Download

Advanced Features

Subfolder Support

You can organize downloads into subfolders:

my-loras/character-lora.safetensors
sdxl/checkpoints/my-model.safetensors

File Override Protection

• If a file already exists, you'll be prompted for confirmation
• Choose to override or cancel the download

Missing Model Detection

• The extension scans your workflow for missing models
• Missing models are highlighted in the UI for quick downloading
• The Download Missing button queues every missing model that already has a URL and folder configured
• If any missing model does not have a URL or folder, the batch action stops immediately and shows which entries need to be fixed first

Example of the batch download controls in the top-right action bar, with the Download Missing button highlighted:

Queue Management

• Multiple downloads are queued automatically
• One download processes at a time for optimal performance
• Cancel queued or active downloads anytime
• Batch downloads reuse the same queue, so one-click actions stay in order with manual downloads

Configuration

The extension shells out to wget for the actual file transfer and polls the output file to report progress back to the UI.

Supported Folders

The downloader supports all standard ComfyUI model directories:

• checkpoints
• loras
• vae
• upscale_models
• embeddings
• controlnet
• clip_vision
• And more...

Security

ComfyUI-Downloader includes multiple security measures:

• Path traversal attack prevention
• Filename validation (no backslashes, dots, or escape sequences)
• Path resolution verification
• Only downloads to configured ComfyUI model directories

API

The extension provides REST API endpoints for programmatic access:

Start Download

POST /{API_PREFIX}/server_download/start
{
  "url": "https://example.com/model.safetensors",
  "save_path": "checkpoints",
  "filename": "model.safetensors",
  "override": false
}

Cancel Download

POST /{API_PREFIX}/server_download/cancel
{
  "download_id": "checkpoints/model.safetensors"
}

WebSocket Events

• server_download_progress: Real-time progress updates
• server_download_complete: Download completion notification
• server_download_error: Error notifications

Development

Project Structure

ComfyUI-Downloader/
├── assets/                # README screenshots
├── __init__.py              # Main extension logic and API endpoints
├── requirements.txt         # Python dependencies
├── web/
│   ├── css/
│   │   └── downloader.css  # UI styling
│   └── js/
│       ├── downloader.js   # Extension initialization
│       └── UI.js           # Modal UI implementation
└── README.md

Building and Testing

1. Make your changes
2. Restart ComfyUI to reload the extension
3. Test in the browser console for any errors

Troubleshooting

Button Doesn't Appear

• Ensure ComfyUI is fully loaded before looking for the button
• Check browser console for JavaScript errors
• Verify the extension is in the custom_nodes directory

Downloads Fail

• Check internet connectivity
• Verify the URL is accessible
• Check browser console and ComfyUI terminal for error messages
• Ensure you have write permissions to the ComfyUI directories
• If Download Missing stops immediately, fill in the missing URL or folder shown in the alert and try again

Slow Download Speeds

• Download performance depends on the remote host and your server network path
• Check that wget is installed and accessible from the ComfyUI process
• Check your network bandwidth and latency

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - see LICENSE file for details

Credits

Created for the ComfyUI community to make model management easier and more efficient.

Support

If you encounter issues or have suggestions:

• Open an issue on GitHub
• Check existing issues for solutions
• Provide error messages and browser console logs when reporting bugs