Prerequisites
- Docker installed on your system (get Docker)
- A terminal or command prompt
What you'll achieve By the end of this guide, you will have TerminusDB running in a Docker container on your machine — ready to accept queries on localhost:6363.
Quick start (2 minutes)
Clone the repository and start the container:
git clone https://github.com/terminusdb/terminusdb
cd terminusdb
docker compose upThat's it. TerminusDB is now running at http://localhost:6363.
Verify it works
Open the dashboard in your browser:
http://127.0.0.1:6363/dashboard/Or test with curl:
curl -u admin:root http://localhost:6363/api/infoYou should see a JSON response with the server version.
What's included
The Docker Compose stack gives you:
| Service | URL | Purpose |
|---|---|---|
| TerminusDB | http://localhost:6363 | Database server + HTTP API |
| Dashboard | http://localhost:6363/dashboard/ | Visual UI for data modelling and queries |
| GraphiQL | http://localhost:6363/api/graphiql/admin/{db} | Interactive GraphQL query browser |
Default credentials: admin / root
Stop and restart
# Stop
docker compose down
# Restart (data persists in Docker volumes)
docker compose upAdding -v to docker compose down -v deletes all data by removing Docker volumes. Only use this if you want a fresh start.
Optional: AI indexing
To enable AI-powered vector indexing, create a .env file before starting:
OPENAI_KEY=your-openai-key-here
BUFFER_AMOUNT=120000The OpenAI key is optional — all database features work without it.
Windows users
On Windows, the default Docker memory is 2 GB. TerminusDB benefits from more memory — increase it in Docker Desktop settings.
For a detailed Windows guide, see Install TerminusDB with Docker on Windows.
Next steps
- Create a Database — your first database in under a minute
- TypeScript Quickstart — connect and start querying
- Python Quickstart — connect with the Python client
- Advanced Docker Configuration — production deployment, CLI access, migration from bootstrap