# Immich

> Immich (116k stars, v3.3.3 Oct 2026) is the closest self-hosted match to Google Photos: Android and iOS apps back up photos and videos in the background, and the web and mobile apps offer a timeline, map, memories, shared and public albums, partner sharing and search by faces, objects and CLIP text. AGPLv3; all features free, with an optional one-time product key to support development. Runs with Docker Compose on Linux; needs 6 GB RAM (4 GB with machine learning off). The default face models come from InsightFace and are licensed for non-commercial research only.

- Repository: https://github.com/immich-app/immich
- Official docs: https://docs.immich.app/
- Checked: 2026-10-16

## What it replaces
- Google Photos: Automatic phone photo and video backup, Search photos by faces, Shared albums and links, Timeline and map browsing (full). Covers backup, search, sharing and timeline/map; you supply the server, storage and backups.
- Dropbox: Mobile camera upload and backup (full). Replaces camera upload only.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Self-hosted with Docker Compose | https://docs.immich.app/install/docker-compose | 2026-10-16 |
| Code licence | Partial | AGPL-3.0 (copyleft) | https://docs.immich.app/FAQ | 2026-10-16 |
| Model weights licence | No | Default facial recognition models (InsightFace buffalo\_l) are for non-commercial research only; ML can be disabled | https://huggingface.co/immich-app/buffalo_l | 2026-10-16 |
| Commercial use | No | AGPL code is fine commercially, but the default InsightFace face models are non-commercial; FUTO also sets commercial guidelines | https://github.com/deepinsight/insightface/tree/master/python-package#license | 2026-10-16 |
| Paid services needed? | Confirmed | None; optional one-time product key supports development and unlocks nothing | https://docs.immich.app/overview/support-the-project | 2026-10-16 |
| Hardware | Confirmed | Minimum 6 GB RAM and 2 cores (8 GB, 4 cores recommended); 4 GB works with machine learning disabled | https://docs.immich.app/install/requirements | 2026-10-16 |
| Operating systems | Confirmed | Server on Linux recommended; Windows and macOS via Docker Desktop (discouraged); Android and iOS apps | https://docs.immich.app/install/requirements | 2026-10-16 |
| Setup effort | Confirmed | Medium; download docker-compose.yml and .env from the release, then docker compose up | https://docs.immich.app/install/docker-compose | 2026-10-16 |

GitHub: 116383 stars, 7227 forks, last push 2026-10-16, archived: no.
---
Source: RepoQuarry (https://repoquarry.dev/projects/immich). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.