marimo

A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Pytho

What is marimo?

A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.

Overview of marimo

What it is

marimo is a reactive Python notebook that aims to replace tools like Jupyter, Streamlit, Jupytext, ipywidgets, and Papermill. Notebooks are stored as pure Python files, making them git-friendly and executable as scripts. The environment is designed for reproducibility: running one cell automatically triggers dependent cells or marks them as stale, keeping code and outputs consistent. marimo also supports first-class SQL queries, AI-native features, and deployment as interactive web apps or slides.

Key features

Reactive execution: run a cell or interact with a UI element, and dependent cells automatically run or are marked staleGit-friendly: stored as .py files instead of JSONBuilt-in SQL engine: query dataframes, databases, lakehouses, CSVs, or Google Sheets; returns results as Python dataframesAI-native: pair with AI agents like Claude Code or use built-in AI assistant with context about variables in memoryReproducible: no hidden state, deterministic execution order, built-in package management with sandboxed venv supportInteractive UI: sliders, dropdowns, dataframe transformers, and chat interfaces automatically re-run dependent codeBatteries-included: includes GitHub Copilot, AI assistants, Ruff code formatting, HTML export, interactive dataframe viewer, VS Code extension, PyCharm plugin

Use cases

marimo is suited for data exploration, reproducible experiments, building interactive dashboards, SQL querying on dataframes and databases, creating shareable web apps or slides from notebooks, and embedding interactive visualizations. It can be used in research, teaching, and data journalism where reproducibility and consistency matter.

Getting started

To install and run the introductory tutorial, use:pip install marimo && marimo tutorial introFor additional dependencies (SQL, AI completion, etc.):pip install "marimo[recommended]"Create notebooks: marimo editRun as web app: marimo run your_notebook.pyExecute as script: python your_notebook.pyConvert Jupyter notebooks: marimo convert your_notebook.ipynb > your_notebook.py

License and pricing

The license is Apache-2.0. marimo is open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/marimo-teammarimo.svg)](https://yunodb.com/tools/marimo-teammarimo?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/marimo-teammarimo?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/marimo-teammarimo.svg" alt="Featured on YunoDB" height="20"></a>