xeronsh/openOii
故事想法 → 多智能体协作 → 漫剧成片 | 基于 LangGraph 的 AI 漫剧生成平台
What it solves
openOii automates the process of turning a story idea into a completed AI-generated comic-style video. It manages the complex pipeline of planning, generating characters and storyboards, and producing and synthesizing the final video, providing a visual way to review the process via an infinite canvas.
How it works
The project uses LangGraph as its core orchestration engine to manage a multi-stage AI generation pipeline. It integrates a FastAPI backend with a React frontend, using WebSockets for real-time progress updates. The system allows users to input story ideas and select styles, then executes a sequence of AI agents to handle planning and asset creation. The results are displayed on a tldraw-powered infinite canvas for review.
Who it’s for
It is primarily designed as a learning and demonstration project for developers interested in multi-stage AI orchestration, state recovery, and real-time frontend-backend collaboration using LangGraph.
Highlights
- Multi-stage Pipeline: Orchestrates the entire flow from story concept to final video synthesis.
- Interactive Canvas: Uses tldraw for an infinite canvas to review characters, storyboards, and final results.
- Execution Control: Supports recoverable, cancellable, and feedback-driven run processes.
- Real-time Updates: Implements WebSocket integration for live progress tracking during generation.
Related
- Project
- Project
- Project
- Project