TagGUI

TagGUI is a desktop app for people preparing image datasets for generative AI training. It combines local AI captioning with manual tag editing, so you can generate descriptions and correct them in the same workspace. It's open source under GPL-3.0 and runs on Windows, Linux and macOS, though macOS doesn't have a packaged release.

Caption generation runs on your hardware, using a CPU or a compatible NVIDIA GPU. The app can download captioning models and use models you've already stored locally. It generates captions or tags for multiple images at once, with controls for prompts, caption prefixes and preferred or discouraged wording. Prompts can draw on an image's existing tags, filename or folder name; wording requests aren't guaranteed to appear in the output.

The editing tools focus on repeated work across a dataset. Keyboard shortcuts and autocomplete based on your most-used tags reduce typing, while batch operations let you rename, delete, sort or reorder tags across images. A built-in Stable Diffusion token counter helps you assess caption length.

Filters let you find images by tags, caption text, filenames or paths, as well as by tag count, character count and token count. You can combine conditions or use wildcards to isolate groups that need attention. TagGUI reads tags from matching text files beside your images and automatically saves edits back to those files.