To get updates like this one delivered to your inbox, subscribe to our newsletter.
This is your monthly reminder that at marimo, we’re on a mission to make the
world’s best programming environment for working with data, and it’s all free
and open source. Just pip install marimo or uv add marimo to get started!
You’re reading the twelfth marimo newsletter.
Since the last newsletter was mailed a month ago, we’ve had many new releases and many new features, including a new utility that makes it easy to embed marimo notebooks in your own blogs or docs.
Before we get into those updates:
- If you haven’t already, check out our YouTube channel for inspiring demos, tutorials, and (soon!) livestreams. Our latest video shows how to make a bulk labeling interface with marimo, featuring UMAP and sentence transformers.
- If you have feedback, we want to hear from you! Either hit reply or head over to https://marimo.io/feedback.
New features
✂️ Announcing marimo snippets
We’ve released marimo snippets, a lightweight utility that lets you embed interactive notebooks directly in your blog posts and documentation, with no JavaScript knowledge required.
All you need to do is load the marimo snippets script on your web page,
and wrap code blocks you’d like to bring to life in <marimo-iframe>
elements. The script does the rest, parsing code snippets and serving
them as a WebAssembly-powered marimo notebook.
To learn more, check out our YouTube video, and to get started visit the snippets repo.
Thanks to Vincent Warmerdam for spearheading snippets!
🤖 Generate marimo notebooks with LLMs
marimo AI is live! Use marimo AI to generate entire notebooks using just English: go from idea to Python (or SQL) notebook in seconds — play with the notebook in our WebAssembly playground‚ or run it locally with the open-source marimo notebook.
To get started, go to http://marimo.app/ai. Then
💬 Enter your own prompt 📚 Try an example from our prompt library
Use generated notebooks as templates to jumpstart data exploration, analysis, modeling, learning — building.
Other updates
- 💻 Bring-your-own (local!) copilots. You can now use custom providers for AI copilots, including your own local models. Check out our docs to learn more.
- Clickhouse support. In just a couple releases, we’ll have built-in support for Clickhouse. This new feature is built by Shahmir Varqha, who joins marimo team from Grab to help build the best user experience seen in any computational notebook.
📬 In case you missed it …
Official videos and blogs from marimo team:
- 🎬 An overview of marimo
- 🎬 How marimo plays well with Git
- 📝 Notebooks as reusable Python programs; this blog made the front page of HackerNews!
🍃 Community
We have over 100 contributors pushing code to marimo, up from 94 last month, 1.6k YouTube subscribers (in just one month!), 1.8k marimonauts hanging out with us on Discord — come chat! We’ve crossed 11.8k stars on GitHub, up from 10.7k last month!
📚 marimo learn
Last month, we announced marimo learn — a curated collection of educational marimo notebooks on foundational concepts related to working with data, spearheaded by prolific marimo team intern Srihari Thyagarajan.
We asked for community contributions, and you all showed up: thank you! Special thanks to @metaboulie, @peter-gy, @jorammutenge, @koushikkhan.
If you’d like to contribute a notebook, we’d love your help! Head over to the repo to learn how.
Roundup
You all are up to so much cool stuff it’s hard to keep up! Here’s our best attempt:
- 📓 Build your own social media timeline algorithm, using this marimo-based blueprint from Mozilla AI scientist Davide Eynard
- 📝 Learn how to implement GRPO, an RL algorithm used by DeepSeek R1, in the latest Hugging Face course, featuring marimo notebooks!
- 📝 “Threat hunters: stop using Jupyter and start using marimo” by Matt Franz
- 🤖 Using marimo to fine-tune a small language model to get better at Rust.
- 📚 A course on data visualization, featuring marimo and anywidget, by Professor Jan Aerts from KU Lueven
- 📓 Play with the Mosaic visualization tool using this WebAssembly notebook by Péter Gyarmati
- 📓 Run semantic search in the browser with Transformers.js and marimo in this notebook by Harry Vanberg
- 📓 Learn how quantile functions work in this tutorial by Stanford scientist Bennet Meyers
Don’t forget to submit your projects to our awesome-marimo repo!
Sincerely,
marimo team 🍃
