Queryable is an open-source iOS and macOS app that lets you search your photo library with natural language, running OpenAI’s CLIP and Apple’s MobileCLIP models entirely on your device. Type a description such as “a brown dog sitting on a bench” and Queryable finds matching photos from your album โ no cloud upload, no account, and no privacy compromise. The project has around 3,000 stars on GitHub and is still actively updated.
Phone photo libraries hold tens of thousands of images, and finding one specific shot by scrolling or by date is painful. Queryable indexes your photos locally with a vision-language model: each photo is embedded once, on-device, and your text queries are matched against those embeddings in the same semantic space. Because Apple Silicon runs MobileCLIP efficiently, the whole experience works offline and stays fast even on large libraries. The app is distributed through the App Store, while the full source code is published on GitHub under the MIT license for anyone to inspect or build themselves. Support for Apple’s MobileCLIP models keeps indexing fast and energy-efficient on modern iPhones, so even very large libraries can be made searchable without noticeable battery impact.
Queryable is free. The source code is open source under the MIT license, and the App Store version is available at no cost, with no subscription or in-app purchase needed for its search features.
Queryable is for iPhone and Mac users with large photo collections who want Google Photos-style semantic search without sending their personal pictures to a cloud service.