Develop with the CLIRun Airflow locally

Run Airflow locally

Running Airflow locally with the Astro CLI lets you preview and debug dag changes before deploying to production. In a local Airflow environment, you can fix issues with your dags without consuming infrastructure resources or waiting on code deploy processes.

The Astro CLI supports multiple modes for running Airflow locally:

  • Container mode (default): Uses Podman or Docker to run Airflow components in containers. All tasks run locally in the scheduler container using the local executor.
  • Standalone mode: Runs Airflow directly on your machine in a virtual environment, without Docker or Podman. This can enable faster iteration in development as you don’t have to wait for containers to rebuild. Use astro dev start --standalone or set it as your default with astro config set dev.mode standalone. See astro dev start for all available options.

See the following documentation to get started: