Personal tutoring,
reimagined.

Agent-native Learning Companion that grows with you.
Fully Open-Sourced.

Current dev Home, with workspace-scoped Home, Learning and Resources and account-level Partners and Personalization.

Eight places, one learning flow.

Chat, Calendar, Notebook, Learning, and Resources follow your active workspace. Partners and Personalization belong to your account, ready to support another subject without mixing its materials with this one.

Workspaces

Separate subjects and projects; keep each workspace’s conversations, learning, and resources together.

Chat

Ask with context, pick a mode, and let one agent loop use your tools, skills, and sources.

III

Calendar

Plan account events beside workspace tasks, due reviews, scheduled instructions, and study activity.

Notebook

Turn sources into your own pages; discuss a passage with Dito and insert evidence from Favorites.

Learning

Study through Books, Mastery Paths, Reading, and Practice, with conversations tied to the activity.

Resources

Reuse chat history, Favorites, questions, learning records, and knowledge bases from this workspace.

VII

Partners

Account-level companions with explicit working context, reusable instructions, and IM channels.

VIII

Personalization

Bring your personas, skills, agents, and connected tools into different workspaces.

Up and running in minutes.

Four ways to install — pick the one that fits. Every path lands at the same configured workspace.

Full Web app + CLI after 1.7.0 publication. Use Source to try the current dev now.

# 1. Choose a runtime-home directory
mkdir -p my-deeptutor && cd my-deeptutor

# 2. Install, configure, launch
pip install -U "deeptutor[full]"
deeptutor init
deeptutor start

Then open 127.0.0.1:3782 Full guide

Published stable image. To run current dev, build the source Dockerfile.

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

Open 127.0.0.1:3782 — the app proxies the API internally. Data persists in the named volume. Full guide

Use the current dev source with the complete Python dependencies and frontend.

git clone --branch dev https://github.com/HKUDS/DeepTutor.git
cd DeepTutor

python3 -m venv .venv && source .venv/bin/activate
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

Python 3.11–3.14 · Node.js 22 LTS. Add --dev to start for frontend hot reload. Full guide

Use the current source from a terminal. The minimal core wheel is a separate installation choice.

git clone --branch dev https://github.com/HKUDS/DeepTutor.git
cd DeepTutor

python3 -m venv .venv-cli && source .venv-cli/bin/activate
python -m pip install -r requirements/full.txt
python -m pip install -e . --no-deps

deeptutor init --cli
deeptutor chat

No frontend setup. Configure embedding when using knowledge bases; check modules before running optional capabilities. Full guide

Personal tutoring, reimagined.