#838 · Primary category: AI Agents & Automation

Pixelle-MCP

An Open-Source Multimodal AIGC Solution based on ComfyUI + MCP + LLM https://pixelle.ai

Project last updated:12/17/25

GitHub Stars

1.1K

Forks

150

Contributors

7

License

MIT

Why we included this project

Pixelle-MCP turns ComfyUI workflows into MCP tools, so any client that speaks the MCP protocol can call them directly. Teams already running ComfyUI locally, or on RunningHub's cloud service, get text, image, sound, and video generation reachable from a chat interface or an autonomous agent without writing glue code. The project also manages the files those workflows produce and presents the results through a Chainlit-based web interface, and a CLI plus PyPI package covers scripted use. LiteLLM integration lets you swap in several model providers as the layer that decides which workflow to run. If your setup already uses MCP clients like Cursor or Claude Desktop, this is a practical way to expose multimodal generation as callable tools.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category