STUDIOYou commented this keyword
Open Generative AI
A free, open-source desktop app for generating AI images and video across many models, powered by the MuAPI service.
- macOS, Windows or Linux
- A MuAPI account for model access
- Setup
- 10 min
- Level
- Beginner
- Stars
- 29.7k
- License
- MIT
What it does
Open Generative AI is a desktop app for making AI images and videos across many models, organized into studios for image, video, audio, lip sync and more. It is the open-source front end; the models run through the MuAPI service behind it.
It is self-hosted on your own machine, and you can add your own models or modify the interface.
Install it
- Go to the Releases page and download the one-click installer for your system (macOS Apple Silicon or Intel, Windows x64, or Linux).
- On macOS, drag the app to Applications. Because it is not notarized, run the Terminal command below once, then right-click the app and choose Open.
- On Windows, if SmartScreen warns you, click More info then Run anyway.
- Open the app and sign in with your MuAPI account to start generating.
macOS first-launch fix
xattr -cr "/Applications/Open Generative AI.app"
Use it
- Prefer not to install anything? Use the hosted version in your browser at muapi.ai/open-generative-ai and sign up for a free account.
- Pick a studio (Image, Video, Lip Sync and so on), choose a model, and describe what you want.
- Some models accept multiple reference images (up to 14 on compatible models).
Good to know
The app is free and open-source, but it runs models through MuAPI, which is a paid service, so generations can cost money. The README describes it as having no content filters, so you are responsible for what you generate and for following the law and MuAPI's terms.
Linux users may need to install the .deb (rather than the AppImage) on Ubuntu 24.04+ due to a sandbox restriction the README explains.
Want the next one first?
One new guide a week, straight to your inbox.
