Logo
Deploy Now

Categories

Knowledge Management Chatbot RAG Education AI Tutoring Learning Platform Personalization

Stars

37,077

Forks

4,645

Watchers

178

Developer links

Last Updated: 08-22-2026 02:59 PM (UTC)

DeepTutor

With 34,000+ GitHub stars and a v1.5 release driven by 36 merged community pull requests, DeepTutor from Hong Kong University's Data Science Lab delivers a full agent-native learning workspace that goes far beyond chatbot wrappers. Eight integrated surfaces — Chat, Deep Solve, Quiz Generation, Deep Research, Math Animator, Co-Writer, Book generation, and Mastery Practice — share a unified context so the objective follows the learner, not the tool. The platform's three-layer memory architecture (L1 working, L2 session, L3 long-term) makes personalization inspectable rather than opaque, letting users see exactly what the system remembers and why. Knowledge retrieval operates across five pluggable engines — LlamaIndex with FAISS vectors, PageIndex for page-level citations, GraphRAG for knowledge-graph traversal, LightRAG for local or server-offloaded retrieval, and linked Obsidian vaults — with document parsing via MinerU, Docling, markitdown, or PyMuPDF4LLM. Partners extend the tutoring brain to 15+ messaging platforms including Slack, Discord, Telegram, Matrix with E2EE, and Mattermost, each carrying private memory with branch, resume, and replay capabilities. Subagent integration brings Claude Code, Codex, Gemini, and Kimi directly into learning sessions. The system supports 30+ LLM providers from OpenAI and Anthropic to Ollama for fully local operation, with multi-user isolation, admin controls, and a full CLI interface. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

DeepTutor
DeepTutor
DeepTutor
DeepTutor
DeepTutor

Benefits

  • Eight Integrated Learning Surfaces
  • Chat, problem-solving, quiz generation, research, visualization, co-writing, book generation, and mastery practice share unified context so learning objectives follow the student across modes.
  • Inspectable Three-Layer Memory
  • Working, session, and long-term memory layers are fully transparent, showing exactly what the system remembers about your learning patterns and knowledge gaps.
  • Five Pluggable RAG Engines
  • Choose LlamaIndex with FAISS, PageIndex for page-level citations, GraphRAG, LightRAG, or linked Obsidian vaults as retrieval backends for your knowledge bases.
  • 30+ LLM Provider Support
  • Connect to OpenAI, Anthropic, DeepSeek, Gemini, Groq, Mistral, Ollama, or any OpenAI-compatible provider without vendor lock-in, including fully local offline operation.

Features

  • TutorBot Partners
  • Persistent IM companions running on 15+ platforms including Slack, Discord, Telegram, Matrix with E2EE, and Mattermost, each with private memory.
  • Quiz Generation
  • Automatically generate quizzes from uploaded course materials with mastery gating, LaTeX rendering, graded questions, and learning progression tracking.
  • Multi-Engine Knowledge Base
  • Upload PDFs, DOCX, XLSX, PPTX files to build versioned retrieval indexes with pluggable document parsing via MinerU, Docling, or PyMuPDF4LLM.
  • Subagent Integration
  • Consult Claude Code, Codex, Gemini, or Kimi live within any chat turn, importing past conversations and maintaining session continuity.
  • Math Animator
  • Visual step-by-step mathematical reasoning with animations that break complex proofs and calculations into digestible progressive visualizations.