What is ComfyUI?

ComfyUI is a powerful and modular GUI and backend that lets you design and execute advanced Stable Diffusion pipelines using a graph/nodes interface, without needing to code.

What models does ComfyUI support?

ComfyUI fully supports SD1.x, SD2.x, SDXL, Stable Video Diffusion, Stable Cascade, SD3, and Stable Audio.

Can ComfyUI run on a computer without a dedicated GPU?

Yes, ComfyUI can run on your CPU by using the --cpu flag when launching the application, although this will be significantly slower than using a GPU.

How do I install ComfyUI?

You can install ComfyUI by either downloading the portable standalone build for Windows, using the Jupyter Notebook, or performing a manual installation on Windows or Linux.

Where do I put my Stable Diffusion checkpoints/models?

Place your Stable Diffusion checkpoints/models (the huge ckpt/safetensors files) in the ComfyUI/models/checkpoints directory.

How do I update ComfyUI?

Updating ComfyUI depends on your installation method. If you used git, you can use git pull from the ComfyUI directory. For other installation methods, refer to the specific documentation.

Can I use ComfyUI offline?

Yes, ComfyUI works fully offline and does not download anything during use.

How can I share models between another UI and ComfyUI?

You can set the search paths for models by modifying the extra_model_paths.yaml file in the ComfyUI directory.

What are some use cases for ComfyUI?

Some use cases include AI-powered image generation, video synthesis, complex workflow design, research & experimentation, and artistic expression.

Where can I find workflow examples for ComfyUI?

You can find workflow examples on the ComfyUI Examples page.

How does ComfyUI integrate with UBOS?

Integrating ComfyUI with UBOS allows you to orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model, and create Multi-Agent Systems.