Self-hosted photo and video management system featuring RAW support, facial recognition, and CLIP visual search.
Store RAW files, run facial recognition, and search image content without third-party cloud services.
Immich is a self-hosted media management tool designed to replace cloud photo services. It gives you full ownership of your media library while providing AI-powered visual search, facial recognition, and automated mobile backups. Photographers needing local control over large RAW archives will find it especially valuable.
Immich ingests photos and videos across mobile and web interfaces, rendering metadata, EXIF details, and interactive map views. It runs local machine learning models like CLIP to enable natural language visual searches and automatic facial clustering. The platform also offers album sharing, photo stacking, user-defined storage hierarchies, and raw format support.
It suits photographers and tech-savvy creators who want complete control over raw storage and metadata on their own hardware. It is not suitable for users who prefer a turnkey cloud subscription without managing server maintenance, databases, or network configurations.
Search your library using natural language content descriptions or query cluster groups built from facial recognition.
Back up and view raw photograph files directly across mobile and web interfaces.
Organize files on disk according to user-defined folder structures rather than locked database paths.
Share entire libraries or specific collections with partners and public guests through secure access links.
Immich is free to install and run under the AGPL-3.0 license. The only direct costs are the hardware, hosting infrastructure, and storage drives you supply yourself.
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
Yes, Immich supports viewing and backing up raw camera formats across both web and mobile platforms.
It uses localized CLIP models and object detection to index visual content so you can search using natural descriptive text.
See Zekai first in Google