SwarmUI
Self-hosted AI image generation UI for Windows, Linux and Apple Silicon Macs. Use Stable Diffusion and Flux with ComfyUI workflows under an MIT license.
SwarmUI is a self-hosted AI image generation interface that combines a straightforward generation screen with direct access to ComfyUI's node-based workflows. It's for people who want to run creative models on their own hardware, with room to build more detailed workflows as their needs grow.
The Generate tab includes an image editor, automatic workflow generation and a Grid Generator for comparing outputs. The Comfy Workflow tab exposes the underlying graph for users who want control over individual processing steps. Both sit in the same browser interface, so you can use the simpler controls and still access workflow editing.
Model support extends beyond images. SwarmUI works with Stable Diffusion, Flux and Krea 2 for image generation, MiniMax H3, Wan and LTX-2 for video, and ACE-Step for audio. It supports ComfyUI and can also use AUTOMATIC1111/stable-diffusion-webui as a backend. Extensions let users add capabilities to the interface.
SwarmUI runs on Windows and Linux, on a self-hosted server, or through Docker. On macOS, it requires an M-series Apple Silicon processor. Multiple GPUs can generate images for the same user at once, which is useful for large comparison grids. You can also run it on rented cloud GPUs through Runpod or Vast.ai; those deployments use remote hardware rather than your own machine.
It's free and open source under the MIT license. Connected backends, extensions and models have their own license terms.
