DeepTutorAI-Powered Personalized Learning

Multi-Agent Architecture for Adaptive Education with Knowledge Graph Integration

📚

Massive Document Q&A

Upload textbooks, papers, and manuals. Build AI-powered knowledge repositories with RAG and knowledge graph integration.

🎨

Interactive Visualization

Transform complex concepts into visual aids with step-by-step breakdowns and engaging interactive demonstrations.

🧠

Smart Problem Solver

Dual-loop reasoning with RAG, web search, paper search, and code execution for step-by-step solutions.

🎯

Practice Generator

Generate targeted quizzes and practice problems. Upload reference exams to mimic original style and difficulty.

🔬

Deep Research

Comprehensive topic exploration with systematic analysis, literature review, and novel insight discovery.

💡

Idea Generation

Automated and interactive concept development with multi-source insights and novelty evaluation.

Why DeepTutor?

DeepTutor transforms how students interact with educational materials through a unified multi-agent architecture. Unlike traditional tools, it provides:

Deep Understanding - Not just answers, but guided learning journeys with visual explanations
Multi-Modal Support - PDF, LaTeX, images, code execution, and more
Knowledge Graph - Semantic connections powered by LightRAG for better comprehension
All-in-One Platform - Problem solving, question generation, research, and idea generation

Quick Demo

bash

# Clone and setup
git clone https://github.com/HKUDS/DeepTutor.git
cd DeepTutor

# Install dependencies
bash scripts/install_all.sh

# Start the application
python scripts/start_web.py