ComfyUI
A free interface for running image and video AI models on your own computer, built by wiring boxes together instead of typing settings.
ComfyUI is the tool most people reach for when they want to run an open image or video model on their own machine. Instead of a single prompt box, it shows a canvas of boxes joined by cables: one box loads the model, another takes your prompt, another does the generating, another saves the file. You drag the connections yourself.
That sounds fiddly, and at first it is. The payoff is that every step is visible and changeable, so you can swap one model for another, add an upscaling step, or reuse the same setup a hundred times without repeating yourself. Nothing leaves your computer, which is the main reason people put up with the learning curve. Expect to need a decent graphics card and an afternoon of patience.