# DeepTutor

> DeepTutor (v1.6.15, Oct 2026, ~47k stars) from the HKU Data Intelligence Lab (HKUDS), described in arXiv 2604.26962, is a learning workspace built on Python and Next.js. It has chat, quiz, research, solve, course study, mastery path, immersive reading and audio overview modes, knowledge bases over your documents (LlamaIndex, LightRAG, GraphRAG and others), editable memory, a task board and a CLI. Run it with Docker or pip install deeptutor, then add an LLM provider in Settings; Ollama, LM Studio and vLLM work for local models. No hosted version is documented. Apache-2.0.

- Repository: https://github.com/HKUDS/DeepTutor
- Official docs: https://deeptutor.info/
- Checked: 2026-10-28

## What it replaces
Nothing people pay for: a free tool, listed because people search for it.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Runs on your own machine or server with Docker, a PyPI install or a source checkout; the web app is served at 127.0.0.1:3782 | https://github.com/HKUDS/DeepTutor#readme | 2026-10-28 |
| Code licence | Confirmed | Apache-2.0 | https://github.com/HKUDS/DeepTutor/blob/main/LICENSE | 2026-10-28 |
| Model weights licence | Confirmed | No bundled model weights; you connect your own LLM and embedding providers | https://github.com/HKUDS/DeepTutor#readme | 2026-10-28 |
| Commercial use | Confirmed | Apache-2.0 allows commercial use; terms of whichever model provider you connect also apply | https://github.com/HKUDS/DeepTutor/blob/main/LICENSE | 2026-10-28 |
| Paid services needed? | Partial | Needs an LLM provider; cloud APIs (OpenAI, Anthropic, Gemini and others) are paid per use, or run free local models through Ollama, LM Studio, llama.cpp, vLLM or Lemonade | https://github.com/HKUDS/DeepTutor#readme | 2026-10-28 |
| Hardware | Unknown | Not documented | none found | - |
| Operating systems | Confirmed | Windows, macOS and Linux via Python 3.11-3.14 (pip or source, with venv steps for both shells), or the Docker image | https://github.com/HKUDS/DeepTutor#readme | 2026-10-28 |
| Setup effort | Confirmed | Medium; one docker run command or pip install deeptutor plus Node.js 20+, then deeptutor init to set the LLM provider and API key | https://github.com/HKUDS/DeepTutor#readme | 2026-10-28 |

GitHub: 54320 stars, 6645 forks, last push 2026-10-31, archived: no.

## Alternatives to DeepTutor
- [Claude How To](https://repoquarry.dev/projects/claude-howto.md): Visual guide to Claude Code in 10 tutorial modules, with Mermaid diagrams and copy-paste slash commands, skills, hooks, MCP and subagent configs.
- [OpenMAIC](https://repoquarry.dev/projects/openmaic.md): Self-hosted AI classroom that turns a topic or document into slides, quizzes and simulations taught by AI teachers and classmates.
- [AI Engineering Hub](https://repoquarry.dev/projects/ai-engineering-hub.md): A collection of 90+ hands-on AI projects and tutorials covering LLM apps, RAG, AI agents, MCP servers and fine-tuning, mostly in Python.
- [Marble Skill Taxonomy](https://repoquarry.dev/projects/os-taxonomy.md): Open JSON dataset of 1,845 primary-school micro-topics linked by 3,941 prerequisite edges and mapped to curriculum standards such as Common Core and NGSS.
- [Claudemy](https://repoquarry.dev/projects/claudemy.md): Claude Code skill that researches any topic on the web and builds a course of about 12 two-minute audio lessons with cards and cited sources.
- [LLM Course](https://repoquarry.dev/projects/llm-course.md): Free LLM course in three tracks (fundamentals, LLM scientist, LLM engineer) with curated reading lists and Colab notebooks for fine-tuning and quantization.
---
Source: RepoQuarry (https://repoquarry.dev/projects/deeptutor). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.