Stonewuu /ai-fusion-video

About ai-fusion-video

Rongguang (融光) is a self-hosted workspace for AI short dramas and motion comics. Scripts, storyboards, assets and generated clips live in one place. It runs with Docker, but you bring your own model API keys. Cloud image and video models can't read files on your own machine.

Best for: Docker-comfortable builders who want to run a short drama from script to finished clips on their own server.

• Organizes projects, members and roles for episodic work
• Writes scripts by episode and scene, and passes context to the agent
• Breaks scripts into shots you can edit, with reference images per shot
• Generates images and videos from text or reference material
• Keeps one asset library for reusable images and videos
• Chats with an agent that takes images, video, audio and files

1. 1
   Install Docker Engine and Docker Compose, then run the command below and cd ai-fusion-video

   git clone https://github.com/Stonewuu/ai-fusion-video.git

   Copy
2. 2
   Copy the env file: cp .env.example .env
3. 3
   Start everything: docker compose up -d
4. 4
   Open http://localhost:8080 and follow the wizard to create the admin account
5. 5
   Change the default MySQL and Redis passwords in .env before you put it online