This workflow will help you set up the Janus Captioning Rig with the 7B model in ComfyUI, enabling you to generate detailed and comprehensive captions for your images efficiently.

1. Create Model Folder:In your ComfyUI installation, navigate to ComfyUi/Models and create a folder named Janus-Pro.
2. Clone 7B Repository:Open a terminal, navigate to the Janus-Pro folder, and run:bashKopierengit clone https://huggingface.co/deepseek-ai/Janus-Pro-7B
3. Clone Janus-Pro Node Repository:Still in the Janus-Pro folder, clone the node repository:bashKopierengit clone https://github.com/CY-CHENYUE/ComfyUI-Janus-Pro.git
4. Install Python Requirements:Navigate to the cloned node repository and install necessary dependencies:bashKopierencd ComfyUI-Janus-Pro pip install -r requirements.txt
5. Restart ComfyUI:Ensure all files are downloaded and dependencies installed, then restart ComfyUI to detect the new model and node.
6. Generate Detailed Captions:Open ComfyUI.Select the Janus-Pro 7B model from the Janus-Pro folder.Upload your images.The rig will automatically generate detailed and descriptive captions for each image, capturing intricate elements and nuances.

Notes:

• System Requirements:Ensure your system has at least 12-16 GB of RAM to run the 7B model smoothly.
• Authentication:If prompted during cloning, use a Hugging Face Access Token.
• Updates:To update the model and node, navigate to their respective folders and run:bashKopierengit pull origin main
• Support:Visit the Janus-Pro 7B Repository and the ComfyUI-Janus-Pro Node Repository for assistance.