rclip

rclip is a command-line semantic photo search tool powered by OpenCLIP's top-performing ViT-B/32 model. Search a local photo library with natural-language queries, similar image search, or mixed text and image queries directly from the terminal. It builds on the CLIP architecture introduced by OpenAI.

Usage: cd photos && rclip "search query"

The first time you run rclip in a directory, it extracts features from your images to build the search index. How long this takes depends on your CPU and the number of images you search. It took about a day to process 73,000 photos on a NAS with an old Intel Celeron J3455.

For a detailed demonstration, watch the video: https://www.youtube.com/watch?v=tAJHXOkHidw.

You can also use an image as the query by passing a file path or image URL. You can also combine and subtract image and text queries. Check out the project's README on GitHub for more usage examples: https://github.com/yurijmikhalevich/rclip#readme.