OpenMAIC  by THU-MAIC

AI platform for immersive, multi-agent educational experiences

Created 1 month ago
14,971 stars

Top 3.4% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

OpenMAIC is an open-source AI platform transforming topics or documents into immersive, multi-agent interactive learning experiences. It targets technical users seeking advanced AI-driven education, offering AI teachers and classmates that generate lectures, quizzes, simulations, and project-based learning activities.

How It Works

The platform uses multi-agent orchestration (LangGraph) and a two-stage generation pipeline (outline -> scenes). Its novelty lies in AI agents capable of real-time speech, whiteboard drawing, and dynamic discussions. Integration with OpenClaw enables classroom generation directly from messaging apps.

Quick Start & Requirements

Installation requires cloning the repo and running pnpm install. Prerequisites include Node.js (>= 20), pnpm (>= 10), and LLM API keys (OpenAI, Anthropic, Google Gemini, etc.). Run locally via pnpm dev or Docker (docker compose up --build). A hosted mode is available at open.maic.chat for zero local setup.

Highlighted Details

  • One-Click Generation: Creates full lessons from a topic or materials in minutes.
  • Multi-Agent Interaction: AI teachers/peers lecture, discuss, and engage with voice, whiteboard, and real-time Q&A.
  • Rich Scene Types: Supports interactive slides, quizzes, HTML simulations, and Project-Based Learning (PBL).
  • Export Options: Generates editable .pptx slides and interactive .html pages.
  • OpenClaw Integration: Seamlessly generates/accesses classrooms from messaging apps (Slack, Discord, Telegram).

Maintenance & Community

The project welcomes community contributions. Specific links to community channels or details on notable contributors/sponsorships are not explicitly provided in the README.

Licensing & Compatibility

Licensed under GNU Affero General Public License v3.0 (AGPL-3.0). This strong copyleft license may impose significant obligations on derivative works, especially concerning distribution. For commercial licensing, contact thu_maic@tsinghua.edu.cn.

Limitations & Caveats

The AGPL-3.0 license requires careful consideration for commercial use or integration into proprietary software due to its copyleft provisions. The platform relies on external LLM APIs, potentially incurring usage costs. Advanced document parsing (MinerU) is optional and may require separate setup or API access.

Health Check
Last Commit

22 hours ago

Responsiveness

Inactive

Pull Requests (30d)
169
Issues (30d)
206
Star History
15,594 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.