Tool not found.
Back to Home
Self-hosted photo and video manager forked from Immich, adding multi-owner Shared Spaces, global people search, S3 storage, and Google Photos import.
Noodle Gallery is a community fork of Immich that rebases onto every upstream release, so all of Immich's fixes and features land automatically, while adding a substantial set of its own on top. The headline addition is Shared Spaces: multi-owner collaborative timelines with Owner/Editor/Viewer roles, cross-contributor face recognition, and a per-space activity log, with photos linked by reference at no extra storage cost — replacing Immich's all-or-nothing Partner Sharing. Global People dedupes face recognition across a user's library and every Shared Space they can access without crossing permission boundaries. Search and discovery gain a command-palette-style Global Search (Cmd/Ctrl+K) with prefix shortcuts for people, tags, albums, and commands, inline structured search filters, timeline grouping by year/month, interdependent filters, and a full filter panel on the map view. AI features extend beyond Immich's faces-and-CLIP baseline with configurable auto-classification of clutter (screenshots, memes, receipts), a dedicated Memories archive with retention controls, location-cluster trip recaps, RF-DETR-based pet detection, and video duplicate detection that averages CLIP embeddings across sampled frames to catch re-encoded duplicates. Media management adds in-browser non-destructive image editing and lossless video trimming (FFmpeg stream copy), the ability to link external photo libraries into Shared Spaces without duplicating files, a guided Google Photos Takeout import wizard, and native S3-compatible storage support (AWS S3, MinIO, R2, B2, Wasabi) with a resumable migration tool. Native iOS and Android apps talk directly to a user's own server. Switching from an existing Immich instance is a three-line Docker config change with full database compatibility, and a one-command script can revert back to upstream Immich cleanly. The project is openly built with heavy AI-assisted development (Claude Code and Codex) alongside human-led design, testing, and debugging.
