ComfyUI — локальная генерация изображений и видео

★ 55k

ComfyUI is an open-source application for generating images and video on your own machine: the model runs on your GPU and there is no per-image fee. Reach for it when you need ad creatives and product shots, covers and illustrations, background or face replacement, upscaling and outpainting, or short clips from a still — and when the same result has to be reproduced dozens of times exactly. The interface is a node graph: every processing step is a block, the wires set the order, and a finished graph can be saved and handed to someone else. That is the key difference from simpler generators — full control over the pipeline at the cost of a slower start. It natively supports current models including Flux, SDXL, SD 3.5, Wan, HunyuanVideo and LTX-Video, plus a large extension ecosystem. A GPU is required; more video memory means larger models fit.

• #UI