RongGuang: Agent-Based End-to-End AI Video Creation Platform 239

What is Rongguang

Rongguang is an open-source, Agent-architecture-based end-to-end AI short-drama creation tool that automates everything from scriptwriting to video generation. It is mainly aimed at content creators: it can automatically break down an input text script into structured storyboards, then call multiple mainstream AI models to generate illustrations and video clips.

Core Features

• Script Management: Supports creating and editing video scripts, with structured organization by episode and scene.

• AI Storyboard Generation: Automatically breaks scripts down into visual storyboards, including visual descriptions and camera language information.

• AI Illustration: Integrates multiple AI illustration engines, enabling one-click generation of storyboard reference images.

• AI Video Generation: Generates corresponding video clips based on storyboard descriptions and reference images.

• Asset Management: Unified management of images, videos, and other assets within a project.

• Multi-Model Support: Integrates mainstream large models such as OpenAI, Claude, Gemini, Tongyi Qianwen, DeepSeek, Ollama, etc.

• Visualized Agent Workflow: Pipeline nodes are displayed visually, making the creation process transparent and controllable.

Technical Architecture

• Backend: Java 21 + Spring Boot 3.5 + Spring AI

• Frontend: Next.js 16 + React 19 + TypeScript

• Database: MySQL + Redis

• Storage: Supports local storage as well as S3-compatible storage such as Alibaba Cloud OSS, Tencent COS, MinIO, etc.

Installation & Usage

Deployment Methods

One-Click Docker Deployment:

1. Install the Docker environment

2. Clone the project repository: git clone https://github.com/Stonewuu/ai-fusion-video.git

3. Run in the project root directory: docker compose up -d

Local Development from Source:

1. Prepare the environment: JDK 21+, Node.js 20+, pnpm 9+, MySQL, Redis

2. Start middleware such as MySQL and Redis first

3. Start the backend and frontend services separately

Usage Flow

1. Create a new project and paste the script or copy

2. Click "AI Smart Breakdown" and the Agent will automatically split the text into structured storyboards

3. Fine-tune prompts and camera-movement parameters in the storyboard editor

4. Execute the workflow; the Agent will automatically schedule: generate reference images → select and lock → generate video clips

5. The system automatically stitches all clips together and exports the final cut

Applicable Scenarios

• Short Video Creation: Batch production for self-media

• Advertising & Marketing: Rapid iteration of multi-version ad storyboards

• Education & Training: Automatic breakdown of teaching outlines into visualized course segments

• Film & TV Pre-visualization: Independent filmmakers quickly generate storyboards and dynamic previews

• Animation Prototyping: Validate narrative pacing and visual style

Rongguang Advantages

• End-to-End Automation: Covers the complete workflow from script to final cut, reducing manual intervention

• Multi-Model Compatibility: Not tied to a single AI provider; supports flexible switching between mainstream large models

• Open Source & Customizable: Supports private deployment and deep feature extensions

• Storyboard-Driven Generation: Uses storyboards as the middle layer to ensure video content aligns with script intent

Comparison with Huobao Short Drama

Github:https://github.com/Stonewuu/ai-fusion-video