Prerequisites
- Python 3.10 or later
- Node.js 18 or later and npm
Step 1: Clone the repository and configure environment
Set up environment variables
.env and fill in your API keys. The required variables are LLM_API_KEY, LLM_MODEL, LLM_HOST, EMBEDDING_API_KEY, EMBEDDING_MODEL, and EMBEDDING_HOST.See Environment variables for a full reference.Step 2: Create a Python environment
- conda (recommended)
- venv
Step 3: Install dependencies
- One-click install (recommended)
- Manual install
On Windows, you may encounter errors about long file paths during
npm install. See the Windows note below.Step 4: Launch DeepTutor
- Frontend: http://localhost:3782
- API docs: http://localhost:8001/docs
Starting services separately
If you need to run the backend and frontend in separate terminals — for example, to view their logs independently — start each one on its own. Backend (FastAPI)web/.env.local so it knows where the backend is:
localhost with the server’s IP address:
Default ports
| Service | Default port |
|---|---|
| Backend (FastAPI) | 8001 |
| Frontend (Next.js) | 3782 |
BACKEND_PORT and FRONTEND_PORT in your .env file before launching.
Windows: long path names
On Windows,npm install can fail with “The filename or extension is too long” due to the default 260-character path limit. Enable long path support by running the following command in an Administrator Command Prompt: