Swayingleaves /novanova-studio

About novanova-studio

NovaNova Studio is a self-hosted workspace for AI image and video work. An agent (an AI that picks tools and runs steps) works on an infinite canvas. Your prompts, assets and results stay in one place. Bring your own paid API keys and object storage.

Best for: Visual teams that want a self-hosted canvas for AI image, video and comic drama work.

• Keeps text, images, videos and results as connected nodes on an infinite canvas
• Generates and edits images and videos in one chat, reusing past results
• Ships agent profiles for image, video, canvas and storyboard work
• Connects OpenAI, Gemini, Anthropic, Agnes, Seedance, MiniMax and Evolink channels
• Lets admins add their own image or video API with request templates
• Runs long video jobs in a queue and streams live progress to your screen

1. 1
   Copy .env.example to .env and fill in your database, Redis, storage and AI channel settings.
2. 2
   Start PostgreSQL and Redis, then run the backend with the command below.

   mvn -f server/pom.xml spring-boot:run

   Copy
3. 3
   Start the frontend with pnpm dev. Set NEXT_PUBLIC_SERVER_URL first if your server port isn't 8080.
4. 4
   Check the backend at http://127.0.0.1:8080/api/v1/health.
5. 5
   For a server, follow the Docker guide and run the compose stack. Flyway runs the migrations for you.