ComfyUI

ComfyUI is an open-source app for running image generation models such as Stable Diffusion and Flux on your own computer. Instead of one prompt box, it shows a graph of nodes: a model loader, prompt encoders, a sampler, an image decoder and so on, wired together into a workflow.

That layout makes each step visible and lets you save, share and reuse whole workflows as files. People use it for upscaling chains, inpainting, consistent characters, video models and batch jobs that would be fiddly in a simpler interface. A large set of community nodes adds more, though each one is extra code you are choosing to run.

It suits people who want control and do not mind a learning curve. It runs fastest on a graphics card with plenty of video memory, and Apple silicon works for many models. Check each model's license, since some open image models limit commercial use.