
PhotoPrism
An AI-powered, privacy-first photo app for browsing, organizing and sharing pictures and videos, running self-hosted or in the cloud with a PWA interface.
- GitHub stars
- 40k
- Last commit
- yesterday
- Latest release
- 261029-963b58c50
- Self-hosted
- Yes
- Hosted version
- Available

PhotoPrism is a photo management application that uses machine learning to help you browse, organize and share photos and videos. It tags and classifies media automatically, so you can search and rediscover pictures without sorting them by hand. The project says it was built to run wherever you need it without compromising privacy, and its topics place it as a Google Photos alternative.
The web interface is a progressive web app that can be installed on a phone, tablet or desktop home screen and feels like a native app. It handles RAW images and many video formats, and offers search filters that can be combined across many properties. A public demo lets you try it first, with a warning not to upload private material. PhotoPrism is written in Go and uses TensorFlow for image analysis.
You can self-host PhotoPrism or use a cloud-hosted setup. Its license is listed as 'Other' on GitHub, so check the terms if you plan to redistribute it. It suits photographers and families who want a private, searchable photo library on hardware they own.
Key features
- AI-based tagging and classification
- Support for RAW photos and video formats
- Combinable search filters
- Installable progressive web app
- Albums, sharing and browsing by place
- Self-hosted or cloud-hosted






















