ComfyBox provides a no-code, customizable frontend for Stable Diffusion, built on top of ComfyUI. It targets users who want to create tailored image generation interfaces without writing code, leveraging the full power and extensibility of ComfyUI's backend.

How It Works

ComfyBox utilizes a novel UI builder that allows users to construct custom interfaces from basic components. It integrates seamlessly with existing ComfyUI workflows, enabling users to import their .json or .png files and automatically generate corresponding UIs. All ComfyUI custom nodes are supported, ensuring broad compatibility.

Quick Start & Requirements

• Installation: Download the latest release, extract, and run run.bat / run.sh. Alternatively, serve the folder contents with any web server.
• Backend: Requires a running ComfyUI instance started with python main.py --enable-cors-header.
• Prerequisites: Python 3 (on PATH), pnpm, Git.
• Docs: [Not explicitly linked, but implies using ComfyUI documentation for backend setup.]

Highlighted Details

• No-code UI builder for custom interfaces.
• Supports importing existing ComfyUI workflows.
• Full compatibility with ComfyUI custom nodes.
• Features a prompt queue, history, and a mobile-friendly version.

Maintenance & Community

• The project is noted as "still under construction."
• No specific community links (Discord, Slack, etc.) or notable contributors are mentioned in the README.

Licensing & Compatibility

• The README does not specify a license.

Limitations & Caveats

The project is explicitly stated to be under construction, with missing features and potential changes to the saved workflow format, meaning users may lose work.