vscode marimo

Run marimo, directly from VS Code.

Note

This extension requires marimo to be installed on your system: pip install marimo. See the installation guide for more details.

Check out the marimo documentation at https://docs.marimo.io/.

Features

• 🚀 Launch marimo from VS Code, in both "edit mode" and "run mode".
• 💻 View the terminal output of marimo directly in VS Code.
• 🌐 View the marimo browser window directly in VS Code or in your default browser.
• 📥 Export notebooks as: html, markdown, or scripts.
• 📓 Convert Jupyter notebooks to marimo notebooks.
• 🧪 [experimental] Run marimo in VSCode's native notebook

Known Issues

VS Code's embedded browser does not support all native browser features. If you encounter any issues, try opening marimo in your default browser instead. For example, the embedded browser will not support PDF render, audio recording, video recording, and some copy/paste operations.

Experimental Native Notebook

This extension includes an experimental feature to run marimo in VSCode's native notebook interface. This feature lets you use VSCode editors and extensions for writing code while the outputs and visualizations are rendered in a view-only marimo editor. This marimo editor displays outputs, console logs, and UI elements to interact with.

This feature is experimental and may have some limitations. Some known limitations are:

• VSCode automatically includes "Run above" and "Run below" buttons in the notebook toolbar. While these work, they do not make sense with a reactive notebook.
• Notebooks can still be edited even though there may not be an active marimo server. This can be confusing since saving or running will not work.
• For autocomplete to work when using native VSCode notebooks for many packages (including marimo, numpy, and more) you may be required to include a pyproject.toml file at the root of the workspace. marimo's editor gets around this by default but unfortunately, the VSCode's native notebook does not.
• You cannot access many marimo features in the native notebook (and need to use the marimo browser), such as the variable explorer, dependency viewer, grid mode (plus other layouts), and more - so we show the notebook in "Kiosk Mode" which is a read-only view of the outputs and helper panels.
• VSCode's native notebook does not support different string quoting styles (e.g. r""", """, f""", etc.), so we default all markdown cells to use r""".

Python Configuration

To ensure marimo works correctly with your Python environment, you have several options:

Tip

The extension will use the Python interpreter from the Python extension by default. Make sure you have the Python extension installed and configured.

1. Workspace Settings (Recommended) Create or edit .vscode/settings.json in your workspace. You can set the default Python interpreter for your entire workspace, or just for marimo.

   For setting the workspace Python interpreter, you can set:

   {
     "python.defaultInterpreterPath": "${workspaceFolder}/.venv/bin/python"
   }

   For setting the Python interpreter only for marimo, you can set:

   {
     "marimo.pythonPath": "${workspaceFolder}/.venv/bin/python"
   }

   If you set marimo.pythonPath, the extension will use that interpreter with -m marimo to invoke marimo.

2. Global Settings You can also configure these settings globally in VS Code's settings:

   • Set python.defaultInterpreterPath to your preferred Python interpreter
   • Verify that marimo is available in your Python interpreter: /value/of/defaultInterpreterPath -m marimo
   • (Likely not needed) Set marimo.marimoPath to the path of your marimo installation. When set, the extension will use this path directly /path/to/marimo instead of python -m marimo.

3. Virtual Environments If using a virtual environment:

   • Create and activate your virtual environment
   • Install marimo: pip install marimo
   • VS Code should automatically detect the Python interpreter

4. uv projects and package environment sandboxes If you are using uv to manage your Python project (e.g. with a pyproject.toml file). You can run uv add marimo to install marimo in your project's environment. Then update your settings to use:

   {
     "marimo.marimoPath": "uv run marimo",
     "marimo.sandbox": true // optional
   }

5. uvx and package environment sandboxes If you are not creating Python projects and don't want to create virtual environments, you can use uvx with marimo edit --sandbox to run marimo in a sandbox.

   {
     "marimo.marimoPath": "uvx marimo",
     "marimo.sandbox": true
   }

Configuration

• marimo.browserType: Browser to open marimo app (system or embedded, default: embedded)
• marimo.port: Default port for marimo server (default: 2818)
• marimo.sandbox: Always start marimo in a sandbox, e.g. marimo edit --sandbox (default: false). Requires uv to be installed.
• marimo.watch: Always start marimo with the --watch flag (default: true).
• marimo.host: Hostname for marimo server (default: localhost)
• marimo.https: Enable HTTPS for marimo server (default: false)
• marimo.enableToken: Enable token authentication (default: false)
• marimo.tokenPassword: Token password (default: empty)
• marimo.showTerminal: Open the terminal when the server starts (default: false)
• marimo.debug: Enable debug logging (default: false)
• marimo.pythonPath: Path to python interpreter (default: the one from python extension). Will be used with /path/to/python -m marimo to invoke marimo.
• marimo.marimoPath: Path to a marimo executable (default: None). This will override use of the pythonPath setting, and instead invoke commands like /path/to/marimo edit instead of python -m marimo edit.

Troubleshooting

If you encounter issues, you can open the marimo extension logs by running the marimo: Show marimo status command from the command palette.

You can also hover over the marimo status bar item in the bottom left of the VSCode window to see the status of the marimo extension.