marimo

What it is and what it does

Marimo is a reactive notebook environment designed as an alternative to traditional notebooks. Unlike Jupyter, marimo automatically tracks cell dependencies and re-runs affected cells when you modify code or interact with UI elements, eliminating hidden state and keeping outputs consistent with code. Notebooks are stored as pure Python files (executable as scripts), can be deployed as interactive web apps, and support SQL queries, dataframe filtering, and AI-assisted code generation.

The package bundles a modern editor with code completion via jedi, syntax highlighting with pygments, and interactive UI elements (sliders, dropdowns, tables, chat interfaces) that bind directly to Python without callbacks. It includes built-in package management, deterministic execution based on variable references rather than cell order, and support for deployment via web services using starlette and uvicorn.

Use it for:

• Build reproducible data analysis workflows that auto-update when inputs change, replacing manual cell re-running
• Deploy interactive dashboards and parameter-driven reports as web apps without rewriting code for a web framework
• Convert notebooks to version-control-friendly Python files that can be executed as scripts or tested with pytest
• Create AI-assisted notebooks that generate cells with context about in-memory variables and data schemas
• Query databases or dataframes with SQL while keeping results synchronized to Python variables

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

A reactive Python notebook environment that automatically runs dependent cells when code changes, stores notebooks as pure Python files, and deploys as scripts or web apps.

Yes. Marimo is actively maintained with no known vulnerabilities, and offers a genuine alternative with reactive semantics that solve real problems (hidden state, manual re-running). Low install friction, permissive Apache-2.0 license, and broad Python version support (3.10–3.14) make it straightforward to adopt. Install if you want reproducible, git-friendly notebooks with automatic dependency tracking; skip if you need deep ecosystem integration with existing .ipynb workflows.

Install

marimo on PyPI

pip install marimo

uv add marimo

poetry add marimo

Installing marimo

Before you install

Low friction installation with a pure wheel distribution. Active maintenance with recent releases; last commit 2026-08-14. Supports Python 3.10 through 3.14. Dependency chain includes web framework (starlette, uvicorn), data processing (narwhals), and editor support (jedi) — all common, well-maintained packages.

License in practice

Licensed under Apache-2.0 (permissive), allowing use in commercial and proprietary projects with minimal restrictions. Derivative works and modifications are permitted provided license and copyright notices are retained.

Quickstart

pip install marimo
marimo edit

# Or run as a script:
python your_notebook.py

# Or deploy as a web app:
marimo run your_notebook.py

Requires Python 3.10 or later.

Verify before relying

• Whether the reactive execution model handles all edge cases in complex dependency graphs without user intervention
• Performance characteristics when working with very large dataframes or long-running cell computations
• Compatibility of SQL query engine with all advertised database and warehouse backends

Package facts

Evidence: marimo-0.23.16-py3-none-any.whl

Tags

More Application Frameworks packages