The box under the desk, now a photo library.

Your NAS already holds your files. Frameleaf turns it into a home for every photo and video, with apps for every phone in the house.

At home in Synology.

A native app, built for your DiskStation. Install Frameleaf Manager through Package Center and give your library a home without leaving DSM.

Frameleaf Manager in a native Synology DSM window, showing library health, storage, services and recent activity.
Your library, right inside DSM. See how your NAS and library are doing, then open your photos. Shown with an example library.
Synology setup assistant with choices to create a library, import Immich or restore a database backup.
A guided start. Create a new library, bring your Immich collection or restore a database backup.
Native Synology service controls for Frameleaf and PostgreSQL, with start, stop, restart and log actions.
Easy to look after. Check services and logs, manage updates and protect your database from one familiar window.

Runs on the NAS you already own.

Give your collection a home on supported hardware. Synology has a native app in Package Center; Unraid, TrueNAS and HexOS have their own setup guides. Check memory, processor and storage requirements first.

Guided setup. Your choice of storage.

Manager checks your storage choices and guides installation, updates and database recovery. Prefer Compose? Use the files supplied together in the published release and configure your folders and password before starting.

Manual container guide

# Get the Compose file and settings from the latest release
wget -O docker-compose.yml https://github.com/Frameleaf/frameleaf-app/releases/latest/download/docker-compose.yml
wget -O .env https://github.com/Frameleaf/frameleaf-app/releases/latest/download/example.env

# Choose storage paths and set a private database password
nano .env

# Start the matching release containers
docker compose up -d

What to plan for.

Memory

6 GB minimum, 8 GB recommended. With 4 GB, run without machine learning.

Processor

Two cores minimum, four recommended. amd64 or arm64.

Database on local disk

Use a local SSD for the database, separate from network shares. Its size grows with your collection and indexes.

Room for your work

Allow space for originals, previews, edited versions, video outputs and backups. What you need depends on your formats and processing settings.

Optional GPU acceleration

Supported Intel and NVIDIA hardware can speed up compatible video and machine learning jobs. Check the guide for your device.

A second copy

Back up both the database and the complete media collection. Keep an independent copy off the server and practise restoring it.