These are just a few of the 12+ enhancements included in 0.17.0. See the release notes for the full list.

github.com/marimo-team...

marimo now supports Python 3.14. Our required dependencies have compiled wheels, so installation remains easy.

To expedite the release, we're shepherding a maintenance-mode fork of msgspec, a fast serialization and validation library.

github.com/marimo-team...

Shout out to Joaquin Coromina, who implemented our MCP support. Read more at his guest blog: marimo.io/blog/beyond...

marimo now exposes its AI tools through an MCP server endpoint, allowing external AI applications like Claude Code, Cursor, and VS Code to interact with your notebooks.

Try it out on molab: molab.marimo.io/notebooks/n...

In the coming weeks, we will have a lot more to say about how marimo is uniquely well-suited to empower both agents and humans to rapidly create production-ready data apps and pipelines ... see our latest newsletter for a sneak peek!

marimo.io/blog/newsle... (n/n)

Previously, agents like Claude Code could already execute marimo notebooks as scripts to get feedback on the correctness of their code.

With marimo check, agents can get additional feedback without running a single line of code, empowering them to self-correct on the fly. (5/n)

Developers and agents prefer marimo to Jupyter because it's code-first and productionizable ...

... while agents wielding marimo make Python accessible to non-technical users who were previously limited by Looker's inflexible UI (4/n)

Companies large and small are replacing legacy data tools with AI-native, workflows built on marimo because of our code-first and productionizable design

With agents, marimo lets companies go instantly from prompt to data app, pipeline, LLM eval notebook, and more (3/n)

marimo.io/blog/marimo... (2/n)