DeepTutor
NewFree📚 Education
Open-source multi-agent AI tutoring system from the University of Hong Kong that transforms static PDF materials into interactive lessons with auto-generated quizzes, concept maps, and adaptive explanations. It represents a research-backed approach to AI education, giving students and self-learners a powerful tool to convert any document into a structured learning experience.
Getting Started with DeepTutor
Step-by-step setup guide
- 1Ensure Python 3.9+ and Git are installed. Clone: git clone https://github.com/HKUDS/DeepTutor
- 2Enter directory: cd DeepTutor && pip install -r requirements.txt
- 3Configure LLM API key (OpenAI, Anthropic): cp .env.example .env
- 4Upload PDF textbooks to the data directory and run the data processing script.
- 5Launch: python app.py starts the local service — open the browser for an interactive learning interface.
Key Features
What DeepTutor offers
PDF Textbook Conversion
Upload PDF textbooks — AI extracts content and generates interactive courses with knowledge structure diagrams and explanations.
Multi-Agent Tutoring
Multiple AI agents collaborate: one explains concepts, another generates quizzes, a third answers questions.
Mind Map Generation
Automatically extracts key concepts and relationships from materials into visual knowledge mind maps.
Adaptive Quizzing
Quiz difficulty adjusts based on your performance, targeting weak areas for focused reinforcement.