Self-hosted AI video and reel generation. One container, one port.
2.8K
Self-hosted AI video and reel generation. Everything runs in one container on your own computer: the app, its database, media storage, and the background job runner. You only need Docker Desktop.
Step-by-step guide: https://napstar-420.github.io/reelcraft/docs/, covering Windows, macOS and Linux.
8080reelcraft-data, container path /dataThe volume holds all your projects, media and settings. Keep using the same volume name, and your work stays safe when you update or recreate the container.
Prefer a terminal? This does the same:
docker run -d --name reelcraft --restart unless-stopped -p 8080:8080 -v reelcraft-data:/data zohaibkhan97/reelcraft
More: running Reelcraft.
Reelcraft works out of the box with a free test provider. Open Settings in its sidebar to:
Keys can also be set as environment variables (OPENROUTER_API_KEY, FAL_KEY,
ELEVENLABS_API_KEY, DEEPGRAM_API_KEY), which take priority over saved ones.
When a new version is out, Reelcraft shows Update to … at the bottom of its sidebar. Click it, then Update. Reelcraft backs up your data, installs the update and restarts itself. If the new version doesn't start, it goes back to the previous one on its own.
Some releases need a newer image. The app tells you when, and the steps are:
reelcraft-data → /data).Database updates run automatically when it starts. Versions before 0.2.0 can't update themselves, so update those this way once. More: updating and backups. Each release's notes are on GitHub Releases.
Reelcraft has no login. Anyone who can reach the port can use it and your API keys. Use it on your own computer or a network you trust, and don't expose port 8080 to the internet.
Content type
Image
Digest
sha256:4e54d50b1…
Size
873.6 MB
Last updated
1 day ago
docker pull zohaibkhan97/reelcraft