Application
AI File Sorter
Privacy-first cross-platform desktop app that sorts, renames, and reviews messy folders with local or remote AI for images, documents, and supported media files.
Gallery
About the project
AI File Sorter is a cross-platform desktop application built in C++ and Qt for Windows, macOS, and Linux. It helps clean up cluttered folders by suggesting categories, folder moves, and better filenames, while keeping the user in control with a review-first workflow before anything is changed.
It can run fully offline with built-in local models such as Gemma 3 4B IT, Gemma 1.1 7B, and Mistral 7B, or use the user's own OpenAI, Gemini, or compatible remote endpoint. Current releases support local visual image analysis, screenshot-aware handling for UI captures, document content analysis for formats such as PDF and Office files, metadata-based rename suggestions for supported audio/video files, category whitelists, More Refined / More Consistent categorization modes, persistent undo, local learning from approved reviews, cache/reset tools, and GPU acceleration via CUDA, Vulkan, or Metal.
The public website is available at filesorter.app, and the source code is published on GitHub.