Skip to content

Getting started


doccano has two options to run:

  • (Recommended) Docker Compose
  • Docker

The usage of docker compose version is explained in the We highly recommend that you should use docker compose version. However, we explain the usage of Docker version and Python/Node version for the additional information.


As a one-time setup, create a Docker container for Doccano:

docker pull doccano/doccano
docker container create --name doccano \
  -e "ADMIN_USERNAME=admin" \
  -e "" \
  -e "ADMIN_PASSWORD=password" \
  -p 8000:8000 doccano/doccano

Next, start Doccano by running the container:

docker container start doccano

To stop the container, run docker container stop doccano -t 5. All data created in the container will persist across restarts.

Go to

For Developers

You can setup local development environment as follows:

git clone
cd doccano
docker-compose -f up

Go to

Or, you can setup via Python and Node.js:


git clone
cd doccano
pipenv sync --dev
pipenv shell
cd backend
python migrate
python create_roles
python create_admin --noinput --username "admin" --email "" --password "password"
python runserver

In another terminal, you need to run Celery to use import/export dataset feature:

cd doccano/backend
celery --app=app worker --loglevel=INFO --concurrency=1


cd frontend
yarn install
yarn dev

Go to