ComfyUI-xiaozhuguang
A ComfyUI extension that adds workflow and node-favorite management, theme customization, alignment and grouping tools, plus image, video, audio and large-model nodes including offline text-to-speech and Qwen-VL inference.
ComfyUI-xiaozhuguang is a beautification and enhancement plugin for the ComfyUI node-based interface. It adds node favoriting with search and usage-based sorting, nested workflow folders with drag-and-drop import, multiple color themes with live preview, custom keyboard shortcuts, and grid alignment and visual grouping tools for arranging large workflows.
Beyond interface tools, it adds image nodes for loading, saving, comparing, splitting, merging and smart-cropping images; video nodes for chunked upload, frame extraction and merging; audio nodes including an offline text-to-speech engine with voice cloning; and nodes for loading and running Qwen and Qwen-VL models. Configuration such as favorites, shortcuts and panel widths can be stored on a backend so it follows the user across browsers. The plugin follows ComfyUI's internationalization conventions, with English source names and Chinese translations.