HKUDS/DeepTutorPublic

DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

AI summary: Multilingual AI tutoring platform providing immersive reading, knowledge base integration, and automated learning paths.

Stars
61.1K
+37 today
Forks
7.4K
Watchers
233
Open issues
165
Open PRs
219
Contributors
~349
Commits
4.1K
Branches
13
Views (30d)
8

PythonApache-2.0Created Dec 28, 2025Last push todayLatest release v1.7.4+1.9K stars this week+19.1K this month

Quick answers

What is DeepTutor?
Multilingual AI tutoring platform providing immersive reading, knowledge base integration, and automated learning paths.
What does DeepTutor do?
DeepTutor is a comprehensive AI-powered educational platform designed to act as a personal tutor and learning assistant. It integrates tightly with external knowledge bases, document parsers, and web search providers to ground its educational content in specific, user-provided materials. The system features an immersive reading environment where source documents are displayed alongside the chat, allowing the AI to cite pages directly. It can autonomously generate structured books, quizzes, and learning mastery paths from raw documents or topics. With support for numerous LLM providers, multi-agent collaboration, and a robust Model Context Protocol (MCP) tool ecosystem, it provides a highly customizable learning experience accessible via web UI, CLI, and various messaging platforms.
Who is DeepTutor for?
Students, educators, and lifelong learners seeking a highly customizable, AI-driven learning environment that grounds its teaching in specific source materials. Requires basic familiarity with configuring API keys for LLM providers.
How popular is DeepTutor on GitHub?
HKUDS/DeepTutor has 61,122 stars and 7,388 forks on GitHub, and gained 1,904 stars in the last 7 days.
What license does DeepTutor use?
HKUDS/DeepTutor is released under the Apache-2.0 license.

Star history

since Jul 27, 2026
020K40K60KJul 2026Sep 2026Oct 2026Nov 2026
61.1K stars as of Nov 21, 2026. Measured daily since Jul 27, 2026; GitHub no longer exposes earlier star timestamps.

Update history

5 recorded
  • Nov 21, 2026New release v1.7.4 (previously v1.7.2).
  • Nov 10, 2026New release v1.7.2 (previously v1.7.1).
  • Oct 31, 2026New release v1.7.1 (previously v1.6.15).
  • Oct 21, 2026New release v1.6.15 (previously v1.6.14).
  • Oct 10, 2026New release v1.6.14 (previously v1.6.12).

Contribution activity

commits per day, last 52 weeks
NovDecJanFebMarAprMayJunJulAugSepOctNovMonWedFri2025-11-23: 0 commits2025-11-24: 0 commits2025-11-25: 0 commits2025-11-26: 0 commits2025-11-27: 0 commits2025-11-28: 0 commits2025-11-29: 0 commits2025-11-30: 0 commits2025-12-01: 0 commits2025-12-02: 0 commits2025-12-03: 0 commits2025-12-04: 0 commits2025-12-05: 0 commits2025-12-06: 0 commits2025-12-07: 0 commits2025-12-08: 0 commits2025-12-09: 0 commits2025-12-10: 0 commits2025-12-11: 0 commits2025-12-12: 0 commits2025-12-13: 0 commits2025-12-14: 0 commits2025-12-15: 0 commits2025-12-16: 0 commits2025-12-17: 0 commits2025-12-18: 0 commits2025-12-19: 0 commits2025-12-20: 0 commits2025-12-21: 0 commits2025-12-22: 0 commits2025-12-23: 0 commits2025-12-24: 0 commits2025-12-25: 0 commits2025-12-26: 0 commits2025-12-27: 0 commits2025-12-28: 3 commits2025-12-29: 8 commits2025-12-30: 17 commits2025-12-31: 11 commits2026-01-01: 11 commits2026-01-02: 7 commits2026-01-03: 9 commits2026-01-04: 7 commits2026-01-05: 15 commits2026-01-06: 26 commits2026-01-07: 10 commits2026-01-08: 8 commits2026-01-09: 21 commits2026-01-10: 10 commits2026-01-11: 8 commits2026-01-12: 4 commits2026-01-13: 5 commits2026-01-14: 6 commits2026-01-15: 16 commits2026-01-16: 2 commits2026-01-17: 2 commits2026-01-18: 14 commits2026-01-19: 3 commits2026-01-20: 1 commit2026-01-21: 3 commits2026-01-22: 2 commits2026-01-23: 4 commits2026-01-24: 2 commits2026-01-25: 4 commits2026-01-26: 1 commit2026-01-27: 1 commit2026-01-28: 1 commit2026-01-29: 6 commits2026-01-30: 1 commit2026-01-31: 1 commit2026-02-01: 1 commit2026-02-02: 1 commit2026-02-03: 4 commits2026-02-04: 3 commits2026-02-05: 1 commit2026-02-06: 1 commit2026-02-07: 1 commit2026-02-08: 1 commit2026-02-09: 1 commit2026-02-10: 2 commits2026-02-11: 1 commit2026-02-12: 3 commits2026-02-13: 1 commit2026-02-14: 1 commit2026-02-15: 1 commit2026-02-16: 1 commit2026-02-17: 1 commit2026-02-18: 1 commit2026-02-19: 1 commit2026-02-20: 1 commit2026-02-21: 1 commit2026-02-22: 1 commit2026-02-23: 1 commit2026-02-24: 1 commit2026-02-25: 1 commit2026-02-26: 1 commit2026-02-27: 1 commit2026-02-28: 1 commit2026-03-01: 1 commit2026-03-02: 1 commit2026-03-03: 1 commit2026-03-04: 1 commit2026-03-05: 1 commit2026-03-06: 1 commit2026-03-07: 1 commit2026-03-08: 1 commit2026-03-09: 4 commits2026-03-10: 1 commit2026-03-11: 1 commit2026-03-12: 1 commit2026-03-13: 5 commits2026-03-14: 4 commits2026-03-15: 2 commits2026-03-16: 1 commit2026-03-17: 1 commit2026-03-18: 1 commit2026-03-19: 1 commit2026-03-20: 2 commits2026-03-21: 1 commit2026-03-22: 3 commits2026-03-23: 1 commit2026-03-24: 6 commits2026-03-25: 1 commit2026-03-26: 1 commit2026-03-27: 1 commit2026-03-28: 1 commit2026-03-29: 3 commits2026-03-30: 1 commit2026-03-31: 1 commit2026-04-01: 1 commit2026-04-02: 6 commits2026-04-03: 1 commit2026-04-04: 3 commits2026-04-05: 1 commit2026-04-06: 4 commits2026-04-07: 12 commits2026-04-08: 14 commits2026-04-09: 1 commit2026-04-10: 11 commits2026-04-11: 6 commits2026-04-12: 5 commits2026-04-13: 9 commits2026-04-14: 5 commits2026-04-15: 3 commits2026-04-16: 0 commits2026-04-17: 5 commits2026-04-18: 3 commits2026-04-19: 0 commits2026-04-20: 13 commits2026-04-21: 5 commits2026-04-22: 9 commits2026-04-23: 5 commits2026-04-24: 6 commits2026-04-25: 4 commits2026-04-26: 2 commits2026-04-27: 1 commit2026-04-28: 1 commit2026-04-29: 6 commits2026-04-30: 2 commits2026-05-01: 10 commits2026-05-02: 2 commits2026-05-03: 4 commits2026-05-04: 3 commits2026-05-05: 6 commits2026-05-06: 17 commits2026-05-07: 10 commits2026-05-08: 1 commit2026-05-09: 2 commits2026-05-10: 3 commits2026-05-11: 3 commits2026-05-12: 26 commits2026-05-13: 8 commits2026-05-14: 17 commits2026-05-15: 2 commits2026-05-16: 8 commits2026-05-17: 5 commits2026-05-18: 4 commits2026-05-19: 8 commits2026-05-20: 22 commits2026-05-21: 38 commits2026-05-22: 7 commits2026-05-23: 1 commit2026-05-24: 0 commits2026-05-25: 0 commits2026-05-26: 0 commits2026-05-27: 8 commits2026-05-28: 6 commits2026-05-29: 2 commits2026-05-30: 0 commits2026-05-31: 0 commits2026-06-01: 0 commits2026-06-02: 0 commits2026-06-03: 6 commits2026-06-04: 0 commits2026-06-05: 0 commits2026-06-06: 2 commits2026-06-07: 3 commits2026-06-08: 0 commits2026-06-09: 0 commits2026-06-10: 4 commits2026-06-11: 8 commits2026-06-12: 1 commit2026-06-13: 12 commits2026-06-14: 14 commits2026-06-15: 0 commits2026-06-16: 0 commits2026-06-17: 8 commits2026-06-18: 3 commits2026-06-19: 3 commits2026-06-20: 5 commits2026-06-21: 7 commits2026-06-22: 3 commits2026-06-23: 10 commits2026-06-24: 4 commits2026-06-25: 0 commits2026-06-26: 0 commits2026-06-27: 3 commits2026-06-28: 0 commits2026-06-29: 2 commits2026-06-30: 4 commits2026-07-01: 0 commits2026-07-02: 0 commits2026-07-03: 1 commit2026-07-04: 2 commits2026-07-05: 0 commits2026-07-06: 0 commits2026-07-07: 0 commits2026-07-08: 0 commits2026-07-09: 2 commits2026-07-10: 1 commit2026-07-11: 1 commit2026-07-12: 1 commit2026-07-13: 0 commits2026-07-14: 1 commit2026-07-15: 6 commits2026-07-16: 2 commits2026-07-17: 4 commits2026-07-18: 10 commits2026-07-19: 18 commits2026-07-20: 3 commits2026-07-21: 2 commits2026-07-22: 2 commits2026-07-23: 4 commits2026-07-24: 23 commits2026-07-25: 0 commits2026-07-26: 10 commits2026-07-27: 5 commits2026-07-28: 5 commits2026-07-29: 6 commits2026-07-30: 0 commits2026-07-31: 1 commit2026-08-01: 0 commits2026-08-02: 2 commits2026-08-03: 11 commits2026-08-04: 4 commits2026-08-05: 3 commits2026-08-06: 10 commits2026-08-07: 6 commits2026-08-08: 0 commits2026-08-09: 2 commits2026-08-10: 2 commits2026-08-11: 4 commits2026-08-12: 15 commits2026-08-13: 5 commits2026-08-14: 1 commit2026-08-15: 0 commits2026-08-16: 30 commits2026-08-17: 15 commits2026-08-18: 0 commits2026-08-19: 17 commits2026-08-20: 26 commits2026-08-21: 17 commits2026-08-22: 2 commits2026-08-23: 10 commits2026-08-24: 32 commits2026-08-25: 11 commits2026-08-26: 26 commits2026-08-27: 23 commits2026-08-28: 29 commits2026-08-29: 18 commits2026-08-30: 50 commits2026-08-31: 18 commits2026-09-01: 31 commits2026-09-02: 1 commit2026-09-03: 24 commits2026-09-04: 1 commit2026-09-05: 1 commit2026-09-06: 25 commits2026-09-07: 19 commits2026-09-08: 16 commits2026-09-09: 8 commits2026-09-10: 51 commits2026-09-11: 23 commits2026-09-12: 2 commits2026-09-13: 4 commits2026-09-14: 39 commits2026-09-15: 12 commits2026-09-16: 10 commits2026-09-17: 12 commits2026-09-18: 1 commit2026-09-19: 5 commits2026-09-20: 3 commits2026-09-21: 28 commits2026-09-22: 11 commits2026-09-23: 111 commits2026-09-24: 14 commits2026-09-25: 8 commits2026-09-26: 5 commits2026-09-27: 79 commits2026-09-28: 23 commits2026-09-29: 20 commits2026-09-30: 10 commits2026-10-01: 7 commits2026-10-02: 2 commits2026-10-03: 7 commits2026-10-04: 30 commits2026-10-05: 0 commits2026-10-06: 0 commits2026-10-07: 83 commits2026-10-08: 52 commits2026-10-09: 2 commits2026-10-10: 2 commits2026-10-11: 12 commits2026-10-12: 12 commits2026-10-13: 33 commits2026-10-14: 3 commits2026-10-15: 9 commits2026-10-16: 4 commits2026-10-17: 3 commits2026-10-18: 2 commits2026-10-19: 29 commits2026-10-20: 6 commits2026-10-21: 11 commits2026-10-22: 8 commits2026-10-23: 10 commits2026-10-24: 4 commits2026-10-25: 2 commits2026-10-26: 20 commits2026-10-27: 38 commits2026-10-28: 52 commits2026-10-29: 39 commits2026-10-30: 55 commits2026-10-31: 87 commits2026-11-01: 51 commits2026-11-02: 41 commits2026-11-03: 25 commits2026-11-04: 12 commits2026-11-05: 16 commits2026-11-06: 25 commits2026-11-07: 5 commits2026-11-08: 4 commits2026-11-09: 48 commits2026-11-10: 27 commits2026-11-11: 26 commits2026-11-12: 18 commits2026-11-13: 16 commits2026-11-14: 7 commits2026-11-15: 31 commits2026-11-16: 1 commit2026-11-17: 0 commits2026-11-18: 0 commits2026-11-19: 0 commits2026-11-20: 6 commits2026-11-21: 0 commits
2,886 commits in the last yearLessMore

Signals and awards

derived from tracked data
  • Landmark project

    61,122 stars

  • Very active

    2,886 commits in 52 weeks

  • Community-driven

    ~349 contributors

  • Permissive license

    Apache-2.0

  • Repeat trending

    56 trending appearances

What DeepTutor does

DeepTutor is a comprehensive AI-powered educational platform designed to act as a personal tutor and learning assistant. It integrates tightly with external knowledge bases, document parsers, and web search providers to ground its educational content in specific, user-provided materials. The system features an immersive reading environment where source documents are displayed alongside the chat, allowing the AI to cite pages directly. It can autonomously generate structured books, quizzes, and learning mastery paths from raw documents or topics. With support for numerous LLM providers, multi-agent collaboration, and a robust Model Context Protocol (MCP) tool ecosystem, it provides a highly customizable learning experience accessible via web UI, CLI, and various messaging platforms.

Students, educators, and lifelong learners seeking a highly customizable, AI-driven learning environment that grounds its teaching in specific source materials. Requires basic familiarity with configuring API keys for LLM providers.

  • Immersive reading: Displays source documents side-by-side with the chat interface, enabling the AI to provide grounded, cited explanations.
  • Automated curriculum generation: Compiles complete books, chapters, quizzes, and structured mastery paths from user-provided topics or documents.
  • Extensive provider support: Connects with major LLM services including OpenAI, Anthropic, Gemini, DeepSeek, and various API gateways.
  • Integrated knowledge bases: Utilizes retrieval-augmented generation (RAG) including native LightRAG and GraphRAG to ground answers in personal libraries.
  • Multilingual interface: Offers localized UI and learning experiences across more than 10 languages including English, Chinese, Japanese, and Spanish.
  • Tool ecosystem: Supports Model Context Protocol (MCP) servers, CLI apps, and executable sandboxes to extend the tutor's capabilities.

Where teams use it

Interactive document study

Upload a dense academic PDF and use the immersive reading mode to ask the AI to explain complex concepts with direct page citations.

Custom course creation

Provide a syllabus topic and have the system automatically compile a structured book complete with chapters and review quizzes.

Knowledge base learning

Connect a personal folder of notes or a GitHub repository and let the RAG engine synthesize answers based strictly on that material.

CLI-based learning

Interact with the tutoring agent directly from the terminal using the native CLI for technical instruction and code execution.

README

main branch

DeepTutor

DeepTutor: Lifelong Personalized Tutoring

GitHub Trending Repository of the Day HKUDS/DeepTutor | Trendshift Star History Rank

Download for macOS (Apple Silicon)  Download for macOS (Intel)  Download for Windows (x64)  Docs — deeptutor.info

English  简体中文  繁體中文  日本語  Español  Français  Arabic  Русский  Hindi  Português  Thai  Polski

License GitHub release arXiv Discord Feishu WeChat

Features · Get Started · Explore · CLI · Ecosystem · Community


🤝 We're actively recruiting maintainers! See the Contributing Guide for the roles, requirements and how to apply.

DeepTutor Home and workspace navigation

📦 Releases

[2026.11.20] v1.7.4 — Desktop security update for macOS and Windows. Install the new app to trust the replacement runtime signing key.

[2026.11.16] v1.7.3 — Flexible reading and book editing, richer search and agent connections, clearer learning records, and more reliable model streams and desktop upgrades.

Past releases (more than 1 week ago)

[2026.11.6] v1.7.2 — Complete desktop downloads, selected-text speech, easier model setup, Korean interface preview, NVIDIA NIM embeddings, and more reliable mastery, retrieval, and updates.

[2026.11.1] v1.7.1 — Desktop previews, broader localization, password changes, checked video lessons, Anki and calendar exports, more providers, and wide reliability improvements.

[2026.10.28] v1.7.0 — Modular installation, a refreshed learning workspace, block notebooks with Dito, narrated video lessons, resumable embedding checkpoints, and more reliable tutoring.

[2026.10.19] v1.6.15 — Reviewed learning-path changes, grouped assessments, bounded tutoring memory, reliable chat recovery, sturdier parsing, retrieval comparisons, and native Gemini speech.

[2026.10.8] v1.6.14 — Original-source visual practice, recoverable indexing, unified video Reading, scientific research tools, managed plugins, a learning journal, and provider setup improvements.

[2026.10.4] v1.6.13 — Reading and listening improvements, question-bank practice, resilient document parsing, configurable speech, workspace-visible attachments, Polish, and new providers.

[2026.9.27] v1.6.12 — Workspace KB moves, Kiwix archives, source figures, Task Board, German UI, and chat recovery.

[2026.9.24] v1.6.11 — French and Ukrainian interfaces, reading folders with in-chat selection actions, figures sent to vision models, paged Office previews, and self-syncing knowledge bases.

[2026.9.22] v1.6.10 — Native LightRAG role models, a published index that records and enforces what built it, PDF attachments that follow your parsing engine, visible truncation, and unfiltered provider choices.

[2026.9.21] v1.6.9 — Folder-based learning workspaces, daily practice, redesigned Settings, clearer streaming conversations, persistent usage accounting, and recoverable archives with explicit permanent deletion.

[2026.9.14] v1.6.8 — A recycle bin for deleted chats, search across your whole conversation history, a tool that looks past your knowledge base, and a sweep of fixes for quiet failures.

[2026.9.11] v1.6.7 — A fix release: books that arrived as one empty chapter, quizzes that produced nothing, the model's scratchpad in the text, formulas printed raw, and cards you could not submit.

[2026.9.8] v1.6.6 — A fix release: answers that could not submit, a copy button that lied, connected knowledge bases for partners, Codex sign-in inside Docker, and a 100 KB lighter home route.

[2026.9.6] v1.6.5 — A content workspace you point at any folder, one exec tool for every language, Mastery Path modes that gate its tools, and Settings that grades readiness.

[2026.9.3] v1.6.4 — Faster isolated runtimes, controllable Book generation, source-complete Mastery paths and Chat hand-offs, durable Reading, unified activity UI, recoverable sessions, and explicit per-model API capabilities.

[2026.9.2] v1.6.3 — Breaking front/back-end refactor, strict canonical routes and recoverable streams, plus learner/guardian accounts, grounded Reading, WeKnora, broader parsing, Python 3.14, and DashScope media.

[2026.8.31] v1.6.2 — Immersive YouTube learning, a plugin-driven Visualize catalog, three new agent harnesses, safer reading citations, multi-format MinerU, live Partner channel status, and guided updates.

[2026.8.30] v1.6.1 — One vendor key linked to every service it serves, a task model for background work, Settings as a searchable navigator, a sidebar you arrange, and first-party LightRAG.

[2026.8.27] v1.6.0 — Faithful EPUB reading and annotations, Courses with Little Tutor and Ask Questions, bounded web-source sync, shared Books with private learning state, and Serply/native search.

[2026.8.25] v1.5.17 — Partners each member owns with private conversations and linkable chat accounts, GitHub repos as a knowledge source, Antigravity CLI, browser WeChat QR login, and deeptutor doctor.

[2026.8.22] v1.5.16 — MarginNote 4 libraries you connect and its add-on fills, Book pages that turn again, and tool-call ids, embeddings and temperature limits that stop breaking behind a gateway.

[2026.8.20] v1.5.15 — PageIndex OSS you host yourself with reasoning retrieval, a question bank you can finally file into, third-party tool/capability plugins, and Apache Tika parsing.

[2026.8.19] v1.5.14 — Immersive Reading: a document open beside the thread, cited page by page; DeepTutor configures itself from chat; IMA libraries you browse and write to; a notebook console.

[2026.8.17] v1.5.13 — Books stream while they compile, track your progress, and export to Markdown; a cost estimate before you approve a spine; and home starter suggestions drawn from memory.

[2026.8.13] v1.5.12 — Web search rebuilt with six new providers (Doubao, Bocha, Zhipu, Firecrawl, Qianfan, Aliyun IQS), a LiteParse parsing engine, MCP servers that reconnect on credential change, and CodeBuddy + OrcaRouter.

[2026.8.10] v1.5.11 — Prose around a DSML tool call stops vanishing, a truncated reply continues instead of ending, live memory usage in Settings, and LightRAG indexing off the event loop.

[2026.8.7] v1.5.10 — Every account signs in to its own Codex, model output language becomes its own setting, empty tool calls are rejected instead of retried, and uploads stop blocking the loop.

[2026.8.4] v1.5.9 — Gemini Embedding 2 on its native endpoint, a per-model reasoning effort control, a Novita AI gateway, retrieval roles for queries, and Compose deployments that keep all of data/.

[2026.8.2] v1.5.8 — Memory: a real heap ceiling for the dev server, source installs serve a production build, bounded LLM client and index caches, and a keep-alive fix for stray 500s.

[2026.7.31] v1.5.7 — A per-account MCP Services store, 101 CLI Apps the tutor can run, credentials moved out of the sandbox's reach, and a mobile layout.

[2026.7.29] v1.5.6 — Remote Codex sign-in completes behind an SSH tunnel, generated files get their own card in Activity, non-English languages stop collapsing to Chinese, and book creation no longer times out.

[2026.7.26] v1.5.5 — Sign in with your ChatGPT plan via OpenAI Codex OAuth, an Eden AI provider, knowledge bases that report what they hold, traceable rag citations, and GraphRAG indexing without a workaround.

[2026.7.24] v1.5.4 — Maintenance sweep: the post-answer "generating" stall is gone, IM partners render Markdown tables faithfully, LLM JSON parsing is sturdier, plus quiz, create-KB form, and Math Animator fixes.

[2026.7.24] v1.5.3 — Themeable code blocks, four more coding CLIs in My Agents (Gemini, Kimi, opencode, MiMo), an Atlas Cloud LLM provider, and a broad chat, memory, embedding, and parsing reliability sweep.

[2026.7.19] v1.5.2 — Configurable chat attachment limits, PageIndex retrieval that reasons across your documents via agentic tool calls, broader Anthropic/OpenAI model support, and steadier Book, Knowledge Base, and chat UI.

[2026.7.9] v1.5.1 — Remove a single failed document from a knowledge base — even one stuck in an error state — instead of deleting and rebuilding the whole base.

[2026.7.4] v1.5.0 — LlamaIndex ingestion now honors your Document Parsing engine with multimodal image extraction, Partner & Soul ids stay URL-safe for non-Latin names, and optional RAG extras install cleanly on Python 3.14+.

[2026.6.30] v1.4.15 — A native Mattermost channel for Partners, plus fixes so Guided Learning multiple-choice questions grade correctly and a configured zero chunk overlap is honored.

[2026.6.29] v1.4.14 — Click an assigned partner to chat in one step, Deep Research flags partial reports, LightRAG indexes without MinerU, FAISS handles non-ASCII paths, and PocketBase sessions are isolated per user.

[2026.6.27] v1.4.13 — Partners support non-Latin names and become assignable to users, logos render after login (#599), tiny knowledge bases retrieve reliably, and containers start cleanly under rootless Podman.

[2026.6.24] v1.4.12 — A new LightRAG Server retrieval engine, a lightweight PyMuPDF4LLM parsing engine, and a FAISS vector backend that makes large knowledge-base retrieval dramatically faster.

[2026.6.23] v1.4.11 — Native tool calling on every cloud OpenAI-compatible provider, a redesigned admin Users page, LaTeX in quiz options, an honest session-loading spinner, and configurable container host binding.

[2026.6.21] v1.4.10 — A self-service Profile page with avatars, a rootless-ready container guide with a single-port request-time proxy, and deny-by-default MCP tools for non-admin users.

[2026.6.19] v1.4.9 — Settings polish: Search shows only the fields your provider needs, connection profiles can be renamed and auto-named by provider, and graded Mastery Path questions flow into your Question Bank.

[2026.6.18] v1.4.8 — Connect your own Partners under My Agents and consult them live in chat — answering through their own persona, library and skills — each with its own private memory.

[2026.6.18] v1.4.7 — Connect your local Claude Code / Codex and consult it live mid-turn, My Agents graduates to a top-level /agents, and Partner conversations gain branch / resume / delete with a replayable trace.

[2026.6.17] v1.4.6 — Four-surface consolidation: a Space learning dashboard with importable My Agents and top-level Memory, a Knowledge Center with GraphRAG / PageIndex / LightRAG / linked-KB / Obsidian, opened-up Settings, and per-model capability gating.

[2026.6.14] v1.4.5 — Guided Learning rebuilt on the chat agent loop with a hard per-type mastery gate and a /learning dashboard, a new loop-plugin framework, plus Markdown export / save-to-notebook for Partner conversations.

[2026.6.13] v1.4.4 — Install community skills from ClawHub with deeptutor skill install behind a security gate, plus real in-browser DOCX/XLSX previews for knowledge-base files.

[2026.6.12] v1.4.3 — TutorBot becomes Partners on a production-grade IM pipeline (15 channels, live streaming), Chat moves to a single agent loop, real per-user isolation, and a rebuilt Visualize.

[2026.5.28] v1.4.2 — Stability + polish: Gemini 2.5+ unblocked across Visualize and Chat, auth-routing fix (#485), smooth-streaming chat UX, a Recents sidebar, and Lemonade local-provider support.

[2026.5.27] v1.4.1 — Security + stability: TutorBot tool sandbox locked down, per-user resource isolation, multimodal image fallback, an HTTP/SSE API for TutorBots, and a v1.4.0 chat regression fix.

[2026.5.22] v1.4.0 — GA cut of v1.4: Auto Mode, three-layer Memory, agentic Deep Research / Solve / Question, LlamaIndex RAG refactor, Visualize/Animator merge, and restart-safe turn runtime.

[2026.5.21] v1.4.0-beta — Three-layer Memory workbench (L1/L2/L3), every chat capability rebuilt on a single agentic engine, LlamaIndex-only RAG, and a unified Settings + Capabilities surface.

[2026.5.10] v1.3.10 — Remote Docker CORS recovery, DISABLE_SSL_VERIFY across SDK providers, safer code-block citations, and optional Matrix E2EE add-on.

[2026.5.9] v1.3.9 — TutorBot Zulip and NVIDIA NIM support, safer thinking-model routing, deeptutor start, sidebar tooltips, and session-store parity.

[2026.5.8] v1.3.8 — Optional multi-user deployments with isolated user workspaces, admin grants, auth routes, and scoped runtime access.

[2026.5.4] v1.3.7 — Thinking-model/provider fixes, visible Knowledge index history, and safer Co-Writer clear/template editing.

[2026.5.3] v1.3.6 — Catalog-based model selection for chat and TutorBot, safer RAG re-indexing, OpenAI Responses token-limit fixes, and Skills editor validation.

[2026.5.2] v1.3.5 — Smoother local launch settings, safer RAG queries, cleaner local embedding auth, and Settings dark-mode polish.

[2026.5.1] v1.3.4 — Book page chat persistence and rebuild flows, chat-to-book references, stronger language/reasoning handling, RAG document extraction hardening.

[2026.4.30] v1.3.3 — NVIDIA NIM + Gemini embedding support, unified Space context for chat history/skills/memory, session snapshots, RAG re-index resilience.

[2026.4.29] v1.3.2 — Transparent embedding endpoint URLs, RAG re-index resilience for invalid persisted vectors, memory cleanup for thinking-model output, Deep Solve runtime fix.

[2026.4.28] v1.3.1 — Stability: safer RAG routing & embedding validation, Docker persistence, IME-safe input, Windows/GBK robustness.

[2026.4.27] v1.3.0 — Versioned KB indexes with re-index workflow, rebuilt Knowledge workspace, embedding auto-discovery with new adapters, Space hub.

[2026.4.25] v1.2.5 — Persistent chat attachments with file-preview drawer, attachment-aware capability pipelines, TutorBot Markdown export.

[2026.4.25] v1.2.4 — Text/code/SVG attachments, one-command Setup Tour, Markdown chat export, compact KB management UI.

[2026.4.24] v1.2.3 — Document attachments (PDF/DOCX/XLSX/PPTX), reasoning thinking-block display, Soul template editor, Co-Writer save-to-notebook.

[2026.4.22] v1.2.2 — User-authored Skills system, chat input performance overhaul, TutorBot auto-start, Book Library UI, visualization fullscreen.

[2026.4.21] v1.2.1 — Per-stage token limits, Regenerate response across all entry points, RAG & Gemma compatibility fixes.

[2026.4.20] v1.2.0 — Book Engine "living book" compiler, multi-document Co-Writer, interactive HTML visualizations, Question Bank @-mention.

[2026.4.18] v1.1.2 — Schema-driven Channels tab, RAG single-pipeline consolidation, externalized chat prompts.

[2026.4.17] v1.1.1 — Universal "Answer now", Co-Writer scroll sync, unified settings panel, streaming Stop button.

[2026.4.15] v1.1.0 — LaTeX block math overhaul, LLM diagnostic probe, Docker + local LLM guidance.

[2026.4.14] v1.1.0-beta — Bookmarkable sessions, Snow theme, WebSocket heartbeat & auto-reconnect, embedding registry overhaul.

[2026.4.13] v1.0.3 — Question Notebook with bookmarks & categories, Mermaid in Visualize, embedding mismatch detection, Qwen/vLLM compatibility, LM Studio & llama.cpp support, and Glass theme.

[2026.4.11] v1.0.2 — Search consolidation with SearXNG fallback, provider switch fix, and frontend resource leak fixes.

[2026.4.10] v1.0.1 — Visualize capability (Chart.js/SVG), quiz duplicate prevention, and o4-mini model support.

[2026.4.10] v1.0.0-beta.4 — Embedding progress tracking with rate-limit retry, cross-platform dependency fixes, and MIME validation fix.

[2026.4.8] v1.0.0-beta.3 — Native OpenAI/Anthropic SDK (drop litellm), Windows Math Animator support, robust JSON parsing, and full Chinese i18n.

[2026.4.7] v1.0.0-beta.2 — Hot settings reload, MinerU nested output, WebSocket fix, and Python 3.11+ minimum.

[2026.4.4] v1.0.0-beta.1 — Agent-native architecture rewrite (~200k lines): Tools + Capabilities plugin model, CLI & SDK, TutorBot, Co-Writer, Guided Learning, and persistent memory.

[2026.1.23] v0.6.0 — Session persistence, incremental document upload, flexible RAG pipeline import, and full Chinese localization.

[2026.1.18] v0.5.2 — Docling support for RAG-Anything, logging system optimization, and bug fixes.

[2026.1.15] v0.5.0 — Unified service configuration, RAG pipeline selection per knowledge base, question generation overhaul, and sidebar customization.

[2026.1.9] v0.4.0 — Multi-provider LLM & embedding support, new home page, RAG module decoupling, and environment variable refactor.

[2026.1.5] v0.3.0 — Unified PromptManager architecture, GitHub Actions CI/CD, and pre-built Docker images on GHCR.

[2026.1.2] v0.2.0 — Docker deployment, Next.js 16 & React 19 upgrade, WebSocket security hardening, and critical vulnerability fixes.

✨ v1.7.4 is live. pip install -U "deeptutor[full]" installs the full Web app and CLI; bare deeptutor installs core only. See the upgrade steps below for legacy CLI installations.

📰 News

  • 2026-10-28 🚀 DeepTutor v1.7.0 is here! A fresh look, smoother navigation, and learning tools that work better together.
  • 2026-05-22 🌐 Official docs site live at deeptutor.info — guides, references, and capability tours in one place.
  • 2026-04-10 📄 Our paper is live on arXiv — read the preprint for the design and ideas behind DeepTutor.
  • 2026-01-01 🎊 Happy New Year! Join our Discord, WeChat, or Discussions — let's shape DeepTutor together.
  • 2025-12-29 🎓 DeepTutor is officially released!

✨ Key Features

DeepTutor is an agent-native learning workspace that connects tutoring, research, creation, and practice in one extensible system.

  • One runtime, many learning modes — Chat, Quiz, Research, Visualize, Solve, Mastery Path, Reading, and more share session context while retaining purpose-built workflows.
  • A connected learning workspace — Home, Learning, and Resources connect Chat, Books, Reading, Practice, Task Board, Calendar, and Learning notebook with shared materials, progress, and learning context.
  • Grounded knowledge and reading — Build versioned libraries with LlamaIndex, PageIndex, GraphRAG, or LightRAG, connect external knowledge sources, and learn beside documents, video, and audio with source-linked citations.
  • Partners and subagents — Consult live agents such as Claude Code, Codex, and Grok CLI, import past conversations, or run persistent Partners with their own memory, resources, and IM channels.
  • Inspectable memory — L1 traces, L2 facts, and L3 synthesis make personalization visible and editable, with evidence links you can audit in the Memory Graph.
  • Extensible by design — Add personas, built-in or community skills, MCP services, CLI apps, and generation models through one agent-native runtime.

🚀 Get Started

Choose the installation path that fits how you want to use DeepTutor. Terminal-based installs share one runtime home: private settings live in data/user/settings/ under the launch directory, or under DEEPTUTOR_HOME / deeptutor start --home when set explicitly. After starting the app, configure a provider and model in Settings → Providers and Language models.

Desktop — macOS and Windows (Preview)

Open DeepTutor bundles Python, Node.js, and the full learning workspace, so no terminal or separate runtime installation is needed. Download published desktop builds from the latest GitHub Release, choosing the installer for your computer below.

For development previews, open a successful Desktop Release workflow run and download the matching Artifact. These previews require GitHub sign-in and expire after seven days; published release downloads do not.

Your computer Download target Installer
Mac with Apple Silicon macos-arm64 Open the DMG and drag Open DeepTutor into Applications.
Mac with Intel macos-x64 Open the DMG and drag Open DeepTutor into Applications.
Windows x64 windows-x64 Run the Setup .exe, or extract the portable ZIP and launch Open DeepTutor.exe.

Windows preview builds are unsigned and may show a SmartScreen warning. See the desktop guide for signing, local data, and source builds.

Docker — Fastest Full-App Trial

Run the full Web app without installing Python or Node.js:

docker run --rm --name deeptutor \
  -p 127.0.0.1:3782:3782 \
  -v deeptutor-data:/app/data \
  ghcr.io/hkuds/deeptutor:latest

Open http://127.0.0.1:3782. The named volume preserves settings and local data; only port 3782 is required for the Web app. See the Docker guide for host workspaces, local model services, reverse proxies, and production deployment.

PyPI — Full Web App and CLI

Use Python 3.11–3.14 and Node.js 20.9+. In the modular 1.7.0 release, deeptutor[full] installs the Web app and CLI; bare deeptutor installs core chat and the SDK.

mkdir -p my-deeptutor && cd my-deeptutor
python -m pip install -U "deeptutor[full]"
deeptutor init     # ports + LLM provider + optional embedding/search
deeptutor start    # starts backend + frontend

Open the URL printed in the terminal, by default http://127.0.0.1:3782. Press Ctrl+C to stop both services.

Upgrading from 1.6.15 or earlier: stop DeepTutor, back up the runtime home, and uninstall a legacy deeptutor-cli package before installing the new family. Keep the same DEEPTUTOR_HOME, then run python -m pip check and deeptutor module list. See the PyPI and migration guide for the exact upgrade commands.

Source — Current Dev and Contributors

Use Python 3.11–3.14 and Node.js 22 LTS to match CI and Docker.

git clone --branch dev https://github.com/HKUDS/DeepTutor.git
cd DeepTutor
python3 -m venv .venv && source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements/full.txt
python -m pip install -e . --no-deps
( cd web && npm ci --legacy-peer-deps )
deeptutor init
deeptutor start --dev

Use deeptutor start without --dev for a production build. Optional dependencies are installed from the repository root, for example python scripts/install_extras.py dev,partners,math-animator. See the source installation guide for Conda, extras, and troubleshooting, or the Agent Setup guide to let a terminal-capable agent perform and verify the setup.

CLI — Terminal and Agent Use

For a minimal terminal installation, use the core wheel; no Node.js or Web frontend is required:

mkdir -p my-deeptutor && cd my-deeptutor
python -m pip install -U deeptutor
deeptutor init --cli
deeptutor chat

Core includes Chat, Ask Questions, the CLI, and the SDK; other capabilities and management commands require their matching feature modules. For the current dev source installation and optional dependencies, see the CLI-only guide. The CLI reference below covers one-shot runs and machine-readable output.

Workspaces and advanced setup: see Content Workspaces, Local Data Persistence, Containerization and Sandbox Deployment, Configuration Reference, and Troubleshooting.

📖 Explore DeepTutor

Start with a workspace for each subject or project. Home, Learning, and Resources change with the active workspace; Partners and Personalization remain available across workspaces.

Area Scope What it contains
Home Workspace Chat, Task Board, Calendar, and Learning notebook
Learning Workspace Books, Mastery Path, Immersive Reading, and Practice
Resources Workspace Chat History, Favorites, Learning journal, Question Bank, and Knowledge Center
Partners Account Persistent companions, Partner Groups, resource access, and IM channels
Personalization Account My Agents, Personas, Skills, MCP Services, CLI Apps, Memory, and Settings
🏗️ System architecture
DeepTutor system architecture
🏠 Home — Conversations, Tasks, Notes, and Calendar

Home is where daily work begins: talk to the tutor, organize next actions, and develop your own notes in the current workspace. Calendar provides an account-wide view while retaining the source workspace for study events.

💬 Chat — The Agent Loop You Actually Use

Chat is the default capability and where most work begins. A single thread can talk normally, call tools, ground itself in selected knowledge bases, read attachments, generate images, consult subagents, write notebook records, and continue with the same context across turns.

DeepTutor chat workspace

The tutor calls tools as needed and can pause to ask clarifying questions. Customize page, at the upper right, chooses the visible modes and Skills, MCP, Partners, Preferences, and voice-input controls. Under Starting prompts, add, edit, or remove the titles and prompts in Explain, Solve, Practice, Plan, and Research. Save applies your account preferences; Cancel discards edits, and Restore defaults brings back the original choices.

DeepTutor chat agent loop

User-toggleable tools are brainstorm, web_search, paper_search, zotero_search, reason, and geogebra_analysis — plus imagegen and videogen once you configure the matching generation model. Contextual tools such as rag, kb_files, knowledge_frontier, read_source, read_memory, write_memory, read_skill, load_tools, exec, web_fetch, ask_user, list_favorites, save_favorite, write_note, search_notes, question_bank, github, consult_subagent, workspace_list, workspace_read, workspace_search, workspace_present, and workspace_export mount automatically when the turn has the right context.

Context comes in two kinds: sticky session context (capability, workspace or course, tools, knowledge bases, persona, model and supported reasoning effort, and Reading / Mastery state) persists across turns; references (files, chat history, books, reading sections, notebooks, selected question-bank entries, imported agents) come from the + menu for one turn by default, or every turn when pinned. The voice button only transcribes the current message.

Reasoning-only output limits. The shared agent loop retries once when internal reasoning uses the output budget without an answer or tool action. The default prompt strategy keeps the same budget. To stop immediately, set strategy: stop under capabilities.chat.reasoning_budget_recovery in your runtime settings/agents.yaml. An explicit larger budget can help models that need more room, at the cost of potentially higher latency and usage:

capabilities:
  chat:
    responding:
      max_tokens: 8000
    reasoning_budget_recovery:
      strategy: increase_output_budget
      max_tokens: 16000

The recovery limit must exceed the normal output limit and be at most 131072; choose a value supported by your model. It applies to one recovery call only, retains the existing round and request-timeout bounds, and leaves reasoning effort unchanged. The trace records the policy, old/new limits, effective per-call budgets, and accumulated usage. Visible answer truncation keeps its normal continuation. With the experimental typescript-native runtime, explicit stop or increase_output_budget policies select the shared Python effects host with the TypeScript scheduler before the turn begins; the result reports that compatibility reason.

The Python history builder also enforces its history plan after condensation. If a complete retained exchange still exceeds that budget, the turn stops with history_budget_exceeded before answer generation. Stored messages and tool-call/result pairs remain intact; start a new conversation or shorten the oversized earlier exchange to continue. This history allowance is separate from the model's full context-window capacity.

With the full profile, Home keeps Chat, Ask Questions, Quiz, Visualize, and Video Lesson one click away; Research for cited reports and Solve for worked reasoning sit under More Capabilities. Select text in an answer to open Little Tutor, a focused sidebar thread that stays nested under the source conversation. With at least three sent messages, the left gutter shows a tick for each question when space allows; hover to preview, click to jump back, or use Alt/Option + ↑/↓ to move between questions.

✅ Task Board & Calendar — Organize Daily Learning

Task Board tracks study tasks in To do, In progress, and Done, with notes, drag-and-drop or keyboard-accessible move controls, and a restorable archive. Tasks stay in the current workspace and require no model configuration.

Calendar keeps personal study sessions, repeating events, and reminders together, with .ics export for Apple Calendar, Google Calendar, Outlook, and Thunderbird. Workspace overlays such as task due dates and reviews remain connected to their source workspace.

✍️ Learning Notebook — Pages, Sources, and Dito
Learning notebook pages

The Learning notebook, formerly Co-Writer, organizes your own writing into pages, subpages, and blocks within the active workspace. Import Word documents, arrange a page tree, and insert original records from Favorites as evidence. Pages keep their save status visible and export as Markdown or Word; interactive content becomes static in Word.

Export flashcards to Anki. Notebook pages can include Flash Cards. On a page that has them, open the page menu and choose Export flashcards (Anki). From the notebook review queue, choose Export all for Anki. In Anki, use File → Import and select the downloaded .txt file. Cards land in a deck named DeepTutor::<note title> with front/back fields, tags, and math preserved. Re-importing the same file updates existing cards instead of duplicating them. DeepTutor's own review schedule is not included.

Learning notebook block editor and Dito

Discuss a passage or block with Dito: ask for an explanation, example, rewrite, or practice question. Conversations remain attached to the relevant page and line. Review proposed changes before accepting them, or insert an answer below the line and keep working. Favorites preserve source records; the notebook is where you build your own understanding.

🎓 Learning — Books, Paths, Reading, and Practice

Learning is the current workspace's study area. Create an activity from Learning → New learning, then reopen it from All learning with its progress, conversations, and source context intact.

Learning activities in the current workspace
Activity Use it for What you keep
Book Build a structured, interactive book from selected sources Editable chapters, typed blocks, reading progress, and Page Chat
Mastery Path Break a learning goal into topics and guided study A knowledge-point map, topic conversations, and mastery evidence
Immersive Reading Work beside a document, web page, video, or audio source Source-linked conversations, verified citations, highlights, notes, and progress
Practice Review collected or imported questions Answers, mistakes, recall ratings, and the next review date
📖 Book — Living Books from Your Materials
DeepTutor book library

Book turns selected sources into an interactive living book — not a static PDF, but a reading environment built from typed blocks. A book can start from knowledge bases, Favorites, question banks, or chat history; the creation flow proposes a chapter outline before content is generated, so you review the shape instead of accepting a blind one-shot output.

Book quiz block   Book outline and chapters   Book interactive widget block

Each chapter compiles into editable typed blocks — text, callouts, quizzes, flash cards, timelines, code, figures, interactive HTML, animations, concept graphs, deep dives, and user notes — with its own Page Chat. Review and undo chapter or section edits; insert, move, regenerate, rewrite, or switch blocks; selected passages enter a learning-capture inbox. Progress, bookmarks, quiz attempts, captures, and Page Chat stay private per reader even when an admin book is shared read-only or for collaborative editing; shared deletion stays admin-only. Any book exports to Markdown, long compiles pause and resume, deeptutor book health flags source drift, and deeptutor book refresh-fingerprints refreshes the source baseline.

🗂️ Resources — Materials and Saved Learning Records

Resources is the current workspace's reusable library. Keep original material, saved results, and learning records here so they can be attached to new conversations and activities without mixing projects.

📚 Knowledge Center — Multi-Engine RAG Libraries
DeepTutor Knowledge Center

Knowledge bases are the document collections behind RAG — they ground Chat turns, Learning notebook edits, Book generation, and Partner conversations. What's distinctive is a choice of retrieval engines: LlamaIndex (the default, hybrid vector + BM25 with optional cross-encoder reranking and exact-flat or HNSW FAISS indexes), PageIndex (reasoning retrieval with page-level citations, hosted or self-hosted OSS), GraphRAG and LightRAG (knowledge-graph retrieval), LightRAG Server (retrieval offloaded to an external LightRAG instance you connect over HTTP), WeKnora (retrieval from a knowledge base in your self-hosted deployment, without a local index or document copy), Tencent IMA (a library you curate in IMA — searched, browsed, and written back to over its OpenAPI), MarginNote 4 (your MN4 study data — documents, excerpts, mind-map cards and the links between them — pushed in by the app's Add-on and navigated with dedicated tools), Kiwix (a searchable ZIM archive served over HTTP without local indexing), or a linked Obsidian vault the tutor reads and writes in place. Each KB is bound to one engine.

Create a knowledge base

Migrating an existing Obsidian, Hermes, or Markdown library? See Knowledge migration guide for connected-vault and indexed-copy paths.

Creating a KB, you can create new (upload and index documents), link existing (reuse an index in place), or connect Kiwix (search one served ZIM archive on demand). Choose its storage workspace at creation; an existing KB can move between workspaces after a preview, keeping assignments and saved references. A KB can also track GitHub repositories (repo, branch, glob) or documentation-site URLs (bounded crawl depth and page count, re-synced every 24 h by default); sync hash-diffs added, changed, and removed content so followed documentation stays current without re-uploading, and linked folders pick up new or changed local files on sync. Re-indexing writes a new flat version-N directory and keeps prior ones, so a working index is never destroyed mid-rebuild. A single document can be removed even from an error-state base — dropping a file that failed to parse without a full delete-and-rebuild. Document parsing — Text-only, MinerU, Docling, Tika, markitdown, PyMuPDF4LLM, or LiteParse — and an optional image-description model are chosen in Settings → Knowledge & documents, with local model downloads off by default. Docling can also run in remote mode against a Docling Serve server (no local install or models needed), configured on that page (mode=remote, a server base URL, and an optional API key) or the DOCLING_MODE / DOCLING_API_BASE_URL / DOCLING_API_TOKEN environment variables. Tika is remote-only and points at the Apache Tika server configured on that page. The CLI mirrors the lifecycle with list/info/create/connect-kiwix/add/search/eval/eval-diff/set-default/delete, source add/remove commands, list-sources, and sync.

The built-in LightRAG engine is installed with pip install 'deeptutor[rag-lightrag]'. That extra contains the supported LightRAG SDK but does not install MinerU. Choose MinerU independently in Document Parsing and either configure its cloud mode or install its current local CLI when structured parsing is wanted. MinerU accepts PDF, common raster images, DOCX, PPTX, and XLSX; the legacy magic-pdf command remains PDF-only. Text-only and the other parsing engines do not require MinerU.

Native LightRAG queries and incremental indexing require the embedding configuration recorded by the published index, including the model, dimension, and endpoint identity. If it changes, restore the original configuration or rebuild with the current embedding; indexes without a recorded embedding identity require a rebuild. The knowledge-base detail and index-version views show recovery guidance, while files remain available for viewing and download.

OpenAI-compatible HTTP embedding responses with explicit vector indices are restored to input order. Repeated, missing, or invalid indices are rejected before the vectors reach indexing; index-free proxy responses retain positional ordering. If a gateway previously returned reordered indexed vectors, rebuild affected indexes to replace their old text/vector associations. Existing index files are preserved.

Scoring a knowledge base. With several engines available, "which one retrieves best here?" is an empirical question — so deeptutor kb eval answers it. Write a QA set (one JSON object per line: a query plus the gold passages an ideal retrieval should return), then score the KB with deeptutor kb eval <name> --dataset qa.jsonl --top-k 5 [--mode hybrid] [--save baseline.json]. Each case is retrieved through the same path a chat turn uses, its ranked citations are matched against the gold passages using exact text or lexical overlap (thresholds are tunable with --min-ratio; semantic equivalence is not assessed), and the run reports Recall@k / Precision@k / nDCG@k / MRR / MAP / Hit@k. Metrics are pure functions over the ranking; no model judges the output. Compare saved reports with deeptutor kb eval-diff baseline.json candidate.json for per-metric and per-case changes. Use --rewrite off or --rewrite standalone on kb eval to compare original and rewritten queries; reports retain rewrite provenance. Cases whose search failed (a missing index, bad credentials) carry no score and are reported separately rather than dragging the means down. PageIndex uses reasoning as retrieval and returns no ranked chunks, so it is excluded.

{"query": "Why do transformers scale attention?", "gold": ["Scaling keeps the dot products from growing with the dimension."]}
🗃️ History, Favorites, Journal, and Question Bank
Resources in the active workspace

Chat History reopens conversations with their original learning context. Favorites preserves useful original conversations and results in movable collections with Markdown export, ready to insert into a Learning notebook. The Learning journal records the current mission, session handoff, and confirmed learning records. The Question Bank organizes questions, answers, explanations, and mistakes; select a set for Learning → Practice to review weak areas and schedule repetition.

🤝 Partners — Persistent Companions Across Workspaces

Partners belong to the account rather than one subject workspace. Each companion keeps its own identity, instructions, model policy, relationships, and channels while using either a private library or an existing workspace's resources.

DeepTutor partners workspace
Identity, Workspace Resources, Groups, and Channels

Every inbound Web or IM message becomes a normal ChatOrchestrator turn — a Partner is "a chat that has a personality and a phone number." Its SOUL.md, model selection, tool policy, and library remain editable after creation. A private Partner workspace copies assigned knowledge bases, skills, and Favorites collections; a bound workspace shares its current files and resources instead. Soul, conversations, and relationship memory stay separate in either mode.

DeepTutor partners architecture

Partner Groups bring two or more Partners into one shared conversation with distinct seats, selectable discussion and shared-memory modes, and durable history. Authenticated users keep private Partner sessions and relationship memory even when a companion is shared.

Per-partner IM channel configuration

Schema-driven channels connect companions to Feishu, Telegram, Slack, Discord, DingTalk, QQ/NapCat, WeCom, WhatsApp, Zulip, Mattermost, Matrix, Mochat, and Microsoft Teams, depending on installed extras and credentials. Browser-based setup supports QR onboarding for selected channels. A Partner can also be connected through Personalization → My Agents and consulted from a normal Chat turn.

🧭 Personalization — Reusable Behavior and Connections

Personalization follows the account across workspaces. Choose how the tutor should behave and which agents, skills, services, applications, and memories it may reuse; workspace and conversation selections can narrow that access further.

Account personalization: agents, personas, skills, and connections
🧑‍🚀 My Agents — Consult Live Agents and Import History
DeepTutor My Agents workspace

Connect a live agent — Claude Code, Codex, Grok CLI, Antigravity, Kimi, opencode, MiMo Code, Hermes Agent, OpenClaw, DeepSeek Harness, a configured ACP agent, or a Partner — and consult it from a Chat turn. DeepTutor runs the selected agent and streams its work into the Activity panel through consult_subagent; the Agent chip or @ keeps the connection attached to the session. Local harnesses use their own authentication, working directory, and session storage.

Import past conversations from ChatGPT, Claude Code, and Codex as searchable, resumable references. ChatGPT uses an official conversations.json export; Claude history is selected by project or working directory, and Codex history by date. Reference an imported conversation through + → My Agents without turning it into DeepTutor's own voice.

🎭 Personas, Skills, MCP Services, and CLI Apps

Personas reuse a voice, teaching style, or response policy and can also seed a Partner's soul. Skills package repeatable workflows as SKILL.md playbooks, with account-shared and workspace-local libraries. Import from EduHub browses community skills and installs them through a security gate (see Ecosystem).

Personalization skill library

MCP Services authorize external tools, while CLI Apps expose installed programs from the CLI-Anything catalog through the execution sandbox. Workspace assignments and user grants determine which resources each turn may use.

🧠 Memory — Inspectable Personalization
DeepTutor memory overview

Memory is a file-backed, three-layer system you can read, curate, and audit — deliberately not a hidden vector store. L1 is the workspace mirror plus an append-only event trace (trace/<surface>/<date>.jsonl); L2 is per-surface curated facts (L2/<surface>.md) with references to L1 entities; L3 is cross-surface synthesis (L3/<profile|recent|scope|preferences>.md) that records its contributing L2 surfaces.

DeepTutor memory graph

The Memory Graph shows the whole pyramid — L3 synthesis at the centre, L2 in the middle ring, L1 traces on the outside — with exact L2 → L1 evidence edges and L3 → contributing-surface links. Memory is tracked across chat, notebook, quiz, kb, book, partner, and cowriter surfaces; the consolidator's Update / Audit / Dedup budgets are tuned in Settings → Memory.

⚙️ Settings — One Control Plane
Settings navigator

Settings is the operational control plane, opening on General for interface and model output language. Its searchable navigator links to independent pages: Personal covers Workspaces, Data migration, Appearance, and Usage statistics; Learning & conversation covers starting points, attachments, Video Learning, learner and guardian controls, Learning progress, and Memory; Models & services covers Providers, Language models, Task models, Embedding, Search, Voice, and Multimodal generation; Features & integrations covers tools, capability parameters, Partners & agents, and Knowledge & documents. System holds Network, Runtime status, and About; Archived chats lets you search, restore, or permanently delete archived conversations. Runtime status contains backend health, resident memory, and the Readiness matrix grading capability blockers, warnings, and suggestions. Workspaces keeps topic files and learning state separate, with verified migration and export under Data migration. A provider holds a vendor's address and credential for reuse by its service models; the model pages choose saved providers and configure model names and capabilities. Task models pin a small, fast model for automatic work — from naming conversations and writing starting points to Math Animator and Video Lesson repairs — and resolve to the active default when empty. Voice groups speech synthesis and transcription; Multimodal generation groups image and video models. Partners & agents configures local harnesses and a remote Hermes gateway.

Vietnamese: In Settings → General, choose Tiếng Việt for Interface language and select Apply changes. Model output language is independent: choose Tiếng Việt there for Vietnamese replies, or keep another reply language while using the Vietnamese interface. Reply quality depends on the selected model's Vietnamese support.

NVIDIA NIM embeddings: Select NVIDIA NIM under Settings → Embedding for text retrieval with nvidia/nemotron-3.5-embed-8b. The default endpoint is https://integrate.api.nvidia.com/v1/embeddings; a custom full endpoint is used as entered. Indexing sends input_type=p

(README truncated)

View on GitHub

Recent activity

commits and pull requests

Releases and announcements

92 total
  1. v1.7.4v1.7.4Nov 20, 20261.2K downloads

    # DeepTutor v1.7.4 Release Notes **Release Date:** 2026.11.20 This desktop security update replaces the runtime signing key used by the macOS and Windows apps. On November 20, desktop release credentials were exposed in GitHub Actions artifacts. Those artifacts have been removed. ## What's New The desktop app verifies runtime updates with a new public key. When it finds a saved runtime signed with the retired key, it installs the verified runtime bundled with the new app and backs up the existing workspace before starting it. Tagged desktop builds now require maintainer approval to access release credentials. ## Upgrade Notes - **macOS:** Download the v1.7.4 DMG for your Mac (Apple Silicon or Intel), quit DeepTutor, and replace the old app in Applications. If macOS blocks the old app from launching, install the replacement directly. - **Windows:** Quit DeepTutor and use the v1.7.4 installer or portable ZIP. The Windows installer and portable app use the same new runtime verification key. - Install the desktop app itself to receive the new public key. Earlier desktop versions still trust the retired key. Until you upgrade, avoid importing core updates into those versions. **F

  2. v1.7.3v1.7.3Nov 15, 20269.4K downloads

    # DeepTutor v1.7.3 Release Notes **Release Date:** 2026.11.16 Building on [v1.7.2](past_releases/ver1-7-2.md), v1.7.3 makes reading and book editing more flexible, expands search and agent connections, and improves learning records, model streams, and desktop upgrades. ## What's New ### Read, zoom, and revisit question help The PDF reader adds continuous zoom, text-fit controls, and an image viewer. Annotation updates and multi-material selection are more reliable. Reading quizzes keep answer keys on the server and preserve question-linked help separately from attempts; authorized guardians can review reading progress, quiz records, and help activity. ### Review book changes before applying them Edit generated chapters and sections without rebuilding the whole book. Structure changes and page-chat proposals show a reviewable diff before application, and the most recent unchanged edit can be undone. New empty sections generate when opened; math rendering and page-chat continuation also receive fixes. ### More search providers and connected agents Search adds DeepSeek, SpecStream, and SearchApi, including SearchApi web, news, and scholar modes. Web Fetch gains a selectable D

  3. v1.7.2v1.7.2Nov 6, 202621.1K downloads

    # DeepTutor v1.7.2 Release Notes **Release Date:** 2026.11.06 Building on [v1.7.1](ver1-7-1.md), v1.7.2 adds selected-text read-aloud, easier model setup, and more reliable guided learning, retrieval, and updates. Desktop releases now require Apple Silicon and Intel Mac installers alongside both Windows downloads. ## What's New ### Complete desktop downloads and sturdier Windows support GitHub Releases provide Apple Silicon and Intel DMGs, a Windows installer, and a Windows portable ZIP. The release workflow checks every platform and all versioned downloads and stable aliases. Windows fixes cover session and migration locks, plugin environments, Python executable discovery, non-UTF-8 output, and paths containing spaces. ### Read just the passage you need Select text in Chat, Mastery Path, or Reading and choose **Read aloud** to hear it through your configured speech provider. Chat read-aloud also remembers playback speed per conversation in the current browser. Voice rendering handles escaped percent signs and degree notation more naturally, and Video Lesson's Edge fallback selects a voice matching the lesson language. ### Easier provider setup and broader compatibility Se

  4. v1.7.1v1.7.1Oct 31, 20266.4K downloads

    # DeepTutor v1.7.1 Release Notes **Release Date:** 2026.11.01 Building on [v1.7.0](ver1-7-0.md), v1.7.1 introduces Open DeepTutor desktop previews, broader localization, and substantial improvements to chat, lessons, accounts, providers, and document handling. Existing runtime data stays in place; no manual migration is required. ## What's New ### Open DeepTutor desktop previews Open DeepTutor bundles Python, Node.js, and the full learning workspace into a desktop app for Apple Silicon Macs, Intel Macs, and Windows x64. Preview builds are available from the Desktop Release workflow, with DMG downloads for macOS and Setup or portable ZIP options for Windows. ### More languages and account controls Vietnamese, Japanese, Turkish, and Brazilian Portuguese join the interface and reply-language options, alongside expanded translations, regional date formatting, and right-to-left rendering. Account-scoped learning text controls adjust font size and line spacing. Users can change their passwords from Profile, while accounts still using an admin-assigned temporary password receive a change prompt. ### Chat that keeps your work and sources together Unsent composer drafts survive nav

  5. v1.7.0v1.7.0Oct 28, 2026

    # DeepTutor v1.7.0 Release Notes **Release Date:** 2026.10.28 Building on [v1.6.15](past_releases/ver1-6-15.md), v1.7.0 brings modular installation, a refreshed learning workspace, block-based notebooks with Dito, and narrated video lessons. Full-app upgrades now use `deeptutor[full]`; existing runtime data remains separate from the installed packages. ## What's New ### Install the parts you use `deeptutor` now provides core chat, CLI, and SDK; `deeptutor[full]` selects the Web app and standard feature modules. The `deeptutor-cli` package becomes a compatibility alias. Inspect installed, enabled, and ready states with `deeptutor module list`, then enable, disable, install, or remove modules while the app is stopped. Dependency checks protect required modules, and removal preserves user data. ### A workspace organized around learning Modern and Classic layouts offer different navigation arrangements. Home brings together Chat, the Task Board, Learning notebook, and Calendar; Learning and Resources keep study activities and materials accessible. Customize Chat modes, controls, and starting prompts, or choose which learning methods appear on the New learning page. Conversation

Code frequency

additions and deletions
+267.1K-267.1KWeek of 2025-12-28: +95,930 linesWeek of 2025-12-28: -3,251 linesWeek of 2026-01-04: +41,903 linesWeek of 2026-01-04: -21,790 linesWeek of 2026-01-11: +68,071 linesWeek of 2026-01-11: -23,885 linesWeek of 2026-01-18: +16,919 linesWeek of 2026-01-18: -8,491 linesWeek of 2026-01-25: +3,332 linesWeek of 2026-01-25: -1,087 linesWeek of 2026-02-01: +14,641 linesWeek of 2026-02-01: -43,572 linesWeek of 2026-02-08: +7,497 linesWeek of 2026-02-08: -13,481 linesWeek of 2026-02-15: +56 linesWeek of 2026-02-15: -56 linesWeek of 2026-02-22: +50 linesWeek of 2026-02-22: -50 linesWeek of 2026-03-01: +49 linesWeek of 2026-03-01: -49 linesWeek of 2026-03-08: +171,982 linesWeek of 2026-03-08: -183,998 linesWeek of 2026-03-15: +23,993 linesWeek of 2026-03-15: -3,573 linesWeek of 2026-03-22: +1,837 linesWeek of 2026-03-22: -3,090 linesWeek of 2026-03-29: +3,117 linesWeek of 2026-03-29: -6,092 linesWeek of 2026-04-05: +11,678 linesWeek of 2026-04-05: -2,709 linesWeek of 2026-04-12: +17,903 linesWeek of 2026-04-12: -14,195 linesWeek of 2026-04-19: +52,319 linesWeek of 2026-04-19: -18,532 linesWeek of 2026-04-26: +26,443 linesWeek of 2026-04-26: -9,253 linesWeek of 2026-05-03: +19,007 linesWeek of 2026-05-03: -8,406 linesWeek of 2026-05-10: +23,167 linesWeek of 2026-05-10: -11,571 linesWeek of 2026-05-17: +69,774 linesWeek of 2026-05-17: -36,578 linesWeek of 2026-05-24: +28,542 linesWeek of 2026-05-24: -1,605 linesWeek of 2026-05-31: +129 linesWeek of 2026-05-31: -61 linesWeek of 2026-06-07: +74,574 linesWeek of 2026-06-07: -56,854 linesWeek of 2026-06-14: +61,521 linesWeek of 2026-06-14: -49,800 linesWeek of 2026-06-21: +8,533 linesWeek of 2026-06-21: -2,990 linesWeek of 2026-06-28: +3,373 linesWeek of 2026-06-28: -1,881 linesWeek of 2026-07-05: +643 linesWeek of 2026-07-05: -152 linesWeek of 2026-07-12: +3,516 linesWeek of 2026-07-12: -900 linesWeek of 2026-07-19: +8,215 linesWeek of 2026-07-19: -1,860 linesWeek of 2026-07-26: +38,285 linesWeek of 2026-07-26: -2,762 linesWeek of 2026-08-02: +74,110 linesWeek of 2026-08-02: -1,080 linesWeek of 2026-08-09: +35,587 linesWeek of 2026-08-09: -22,215 linesWeek of 2026-08-16: +70,910 linesWeek of 2026-08-16: -72,768 linesWeek of 2026-08-23: +107,708 linesWeek of 2026-08-23: -18,131 linesWeek of 2026-08-30: +267,140 linesWeek of 2026-08-30: -110,220 linesWeek of 2026-09-06: +81,565 linesWeek of 2026-09-06: -18,675 linesWeek of 2026-09-13: +28,994 linesWeek of 2026-09-13: -4,976 linesWeek of 2026-09-20: +118,831 linesWeek of 2026-09-20: -32,674 linesWeek of 2026-09-27: +34,530 linesWeek of 2026-09-27: -5,565 linesWeek of 2026-10-04: +75,048 linesWeek of 2026-10-04: -19,070 linesWeek of 2026-10-11: +50,846 linesWeek of 2026-10-11: -6,500 linesWeek of 2026-10-18: +55,523 linesWeek of 2026-10-18: -9,193 linesWeek of 2026-10-25: +119,128 linesWeek of 2026-10-25: -51,365 linesWeek of 2026-11-01: +32,047 linesWeek of 2026-11-01: -7,037 linesWeek of 2026-11-08: +39,724 linesWeek of 2026-11-08: -2,040 linesWeek of 2026-11-15: +11,162 linesWeek of 2026-11-15: -1,233 linesDec 28, 2025Nov 15, 2026
+2.1M lines added, -915.3K removed over the last year.

Commits per week

last 52 weeks
2930Week of 2025-11-23: 0 commitsWeek of 2025-11-30: 0 commitsWeek of 2025-12-07: 0 commitsWeek of 2025-12-14: 0 commitsWeek of 2025-12-21: 0 commitsWeek of 2025-12-28: 66 commitsWeek of 2026-01-04: 97 commitsWeek of 2026-01-11: 43 commitsWeek of 2026-01-18: 29 commitsWeek of 2026-01-25: 15 commitsWeek of 2026-02-01: 12 commitsWeek of 2026-02-08: 10 commitsWeek of 2026-02-15: 7 commitsWeek of 2026-02-22: 7 commitsWeek of 2026-03-01: 7 commitsWeek of 2026-03-08: 17 commitsWeek of 2026-03-15: 9 commitsWeek of 2026-03-22: 14 commitsWeek of 2026-03-29: 16 commitsWeek of 2026-04-05: 49 commitsWeek of 2026-04-12: 30 commitsWeek of 2026-04-19: 42 commitsWeek of 2026-04-26: 24 commitsWeek of 2026-05-03: 43 commitsWeek of 2026-05-10: 67 commitsWeek of 2026-05-17: 85 commitsWeek of 2026-05-24: 16 commitsWeek of 2026-05-31: 8 commitsWeek of 2026-06-07: 28 commitsWeek of 2026-06-14: 33 commitsWeek of 2026-06-21: 27 commitsWeek of 2026-06-28: 9 commitsWeek of 2026-07-05: 4 commitsWeek of 2026-07-12: 24 commitsWeek of 2026-07-19: 52 commitsWeek of 2026-07-26: 27 commitsWeek of 2026-08-02: 36 commitsWeek of 2026-08-09: 29 commitsWeek of 2026-08-16: 107 commitsWeek of 2026-08-23: 149 commitsWeek of 2026-08-30: 126 commitsWeek of 2026-09-06: 144 commitsWeek of 2026-09-13: 83 commitsWeek of 2026-09-20: 180 commitsWeek of 2026-09-27: 148 commitsWeek of 2026-10-04: 169 commitsWeek of 2026-10-11: 76 commitsWeek of 2026-10-18: 70 commitsWeek of 2026-10-25: 293 commitsWeek of 2026-11-01: 175 commitsWeek of 2026-11-08: 146 commitsWeek of 2026-11-15: 38 commitsNov 23, 2025Nov 15, 2026
2.9K commits in the last 52 weeks.

When work happens

weekday and hour
SunMonTueWedThuFriSat036912151821Sun 0:00 — 36 commitsSun 1:00 — 35 commitsSun 2:00 — 25 commitsSun 3:00 — 13 commitsSun 4:00 — 23 commitsSun 5:00 — 15 commitsSun 6:00 — 6 commitsSun 7:00 — 6 commitsSun 8:00 — 1 commitsSun 9:00 — 25 commitsSun 10:00 — 40 commitsSun 11:00 — 18 commitsSun 12:00 — 13 commitsSun 13:00 — 16 commitsSun 14:00 — 17 commitsSun 15:00 — 24 commitsSun 16:00 — 19 commitsSun 17:00 — 22 commitsSun 18:00 — 23 commitsSun 19:00 — 20 commitsSun 20:00 — 13 commitsSun 21:00 — 10 commitsSun 22:00 — 15 commitsSun 23:00 — 26 commitsMon 0:00 — 35 commitsMon 1:00 — 42 commitsMon 2:00 — 32 commitsMon 3:00 — 23 commitsMon 4:00 — 6 commitsMon 5:00 — 1 commitsMon 6:00 — 4 commitsMon 7:00 — 7 commitsMon 8:00 — 11 commitsMon 9:00 — 11 commitsMon 10:00 — 24 commitsMon 11:00 — 20 commitsMon 12:00 — 14 commitsMon 13:00 — 12 commitsMon 14:00 — 17 commitsMon 15:00 — 11 commitsMon 16:00 — 18 commitsMon 17:00 — 23 commitsMon 18:00 — 13 commitsMon 19:00 — 19 commitsMon 20:00 — 30 commitsMon 21:00 — 18 commitsMon 22:00 — 18 commitsMon 23:00 — 23 commitsTue 0:00 — 46 commitsTue 1:00 — 20 commitsTue 2:00 — 25 commitsTue 3:00 — 6 commitsTue 4:00 — 0 commitsTue 5:00 — 4 commitsTue 6:00 — 4 commitsTue 7:00 — 4 commitsTue 8:00 — 6 commitsTue 9:00 — 14 commitsTue 10:00 — 11 commitsTue 11:00 — 19 commitsTue 12:00 — 26 commitsTue 13:00 — 15 commitsTue 14:00 — 14 commitsTue 15:00 — 13 commitsTue 16:00 — 18 commitsTue 17:00 — 31 commitsTue 18:00 — 20 commitsTue 19:00 — 18 commitsTue 20:00 — 14 commitsTue 21:00 — 31 commitsTue 22:00 — 14 commitsTue 23:00 — 19 commitsWed 0:00 — 33 commitsWed 1:00 — 22 commitsWed 2:00 — 24 commitsWed 3:00 — 3 commitsWed 4:00 — 2 commitsWed 5:00 — 3 commitsWed 6:00 — 1 commitsWed 7:00 — 2 commitsWed 8:00 — 8 commitsWed 9:00 — 15 commitsWed 10:00 — 64 commitsWed 11:00 — 50 commitsWed 12:00 — 35 commitsWed 13:00 — 12 commitsWed 14:00 — 16 commitsWed 15:00 — 12 commitsWed 16:00 — 16 commitsWed 17:00 — 25 commitsWed 18:00 — 19 commitsWed 19:00 — 18 commitsWed 20:00 — 100 commitsWed 21:00 — 18 commitsWed 22:00 — 29 commitsWed 23:00 — 34 commitsThu 0:00 — 45 commitsThu 1:00 — 40 commitsThu 2:00 — 15 commitsThu 3:00 — 8 commitsThu 4:00 — 5 commitsThu 5:00 — 8 commitsThu 6:00 — 6 commitsThu 7:00 — 11 commitsThu 8:00 — 7 commitsThu 9:00 — 6 commitsThu 10:00 — 6 commitsThu 11:00 — 10 commitsThu 12:00 — 55 commitsThu 13:00 — 21 commitsThu 14:00 — 17 commitsThu 15:00 — 30 commitsThu 16:00 — 17 commitsThu 17:00 — 14 commitsThu 18:00 — 28 commitsThu 19:00 — 28 commitsThu 20:00 — 11 commitsThu 21:00 — 15 commitsThu 22:00 — 29 commitsThu 23:00 — 41 commitsFri 0:00 — 40 commitsFri 1:00 — 18 commitsFri 2:00 — 21 commitsFri 3:00 — 11 commitsFri 4:00 — 8 commitsFri 5:00 — 0 commitsFri 6:00 — 1 commitsFri 7:00 — 10 commitsFri 8:00 — 4 commitsFri 9:00 — 9 commitsFri 10:00 — 8 commitsFri 11:00 — 10 commitsFri 12:00 — 9 commitsFri 13:00 — 21 commitsFri 14:00 — 16 commitsFri 15:00 — 17 commitsFri 16:00 — 34 commitsFri 17:00 — 18 commitsFri 18:00 — 7 commitsFri 19:00 — 15 commitsFri 20:00 — 19 commitsFri 21:00 — 8 commitsFri 22:00 — 23 commitsFri 23:00 — 13 commitsSat 0:00 — 20 commitsSat 1:00 — 21 commitsSat 2:00 — 6 commitsSat 3:00 — 3 commitsSat 4:00 — 5 commitsSat 5:00 — 0 commitsSat 6:00 — 1 commitsSat 7:00 — 1 commitsSat 8:00 — 2 commitsSat 9:00 — 0 commitsSat 10:00 — 6 commitsSat 11:00 — 16 commitsSat 12:00 — 7 commitsSat 13:00 — 7 commitsSat 14:00 — 10 commitsSat 15:00 — 7 commitsSat 16:00 — 22 commitsSat 17:00 — 27 commitsSat 18:00 — 11 commitsSat 19:00 — 7 commitsSat 20:00 — 10 commitsSat 21:00 — 28 commitsSat 22:00 — 21 commitsSat 23:00 — 14 commits
Commit volume by weekday and hour (UTC). Larger dots mean more commits.
DateListRankStars gained
Nov 21, 2026monthly#7+19,312
Nov 20, 2026monthly#7+19,312
Nov 19, 2026monthly#7+19,065
Nov 18, 2026monthly#8+18,753
Nov 17, 2026monthly#6+18,533
Nov 16, 2026monthly#6+18,278
Nov 15, 2026monthly#8+18,077
Nov 14, 2026monthly#7+17,913
Nov 13, 2026monthly#7+17,692
Nov 12, 2026monthly#7+17,406
Nov 11, 2026monthly#7+17,107
Nov 10, 2026monthly#6+16,710
Nov 9, 2026monthly#6+16,065
Nov 8, 2026monthly#9+15,721
Nov 7, 2026weekly#6+6,229