An open-source reactive Python notebook: reproducible, git-friendly, execute as scripts, share as apps!
GitHub: https://github.com/marimo-team/marimo
Discord: https://marimo.io/discord?ref=bsky
@marimo.ioCase study: marimo.io/blog/case-s...
marimo
@marimo.io
@marimo.io@marimo.ioHow do you catch an AI agent that gets the right answer the wrong way? Acto built marimo apps that put experts in front of the evidence, bringing human review back into AI evals. Read the full case study below ⬇️
@marimo.io🧪 Less than 48 hours left in the OpenADMET × marimo cheminformatics notebook competition!
Turn a drug-discovery question into an interactive marimo notebook and compete for $2.5K+ in prizes, including a Mac mini M6.
Submit by October 4, 11:59 PM PT → marimo.io/pages/event...
@marimo.iomarimohub announcement closes our launch week. All announcements, in one place: marimo.io/blog/launch...
Head over to our Reddit and Discord and let us know, what is the most useful for your workflow: pixi, fsspec, lens, studio or the hub?
@marimo.iomarimohub is out, free and open source under Apache 2.0.
Watch Vincent's walkthrough:
www.youtube.com/watch?v=1wn...
@marimo.ioOrganize notebooks into projects, share data connections, and choose who can edit, view, or use an app. Publish interactive apps for teammates, schedule notebook jobs, and review or restore earlier versions—all from the same hub.
@marimo.ioUse the infrastructure that fits your team. Docker on a VM or Kubernetes. S3-compatible storage or a filesystem. Your existing identity provider. Notebook files, versions, and metadata live in your storage backend. The notebooks stay ordinary Python files.
@marimo.ioIntroducing marimohub 🌊
Your team's notebooks. Your infrastructure.
An open-source platform to run and share marimo notebooks with your own compute, storage, and SSO. No separate database to operate.
Read @themylesfiles' announcement and get started: marimo.io/blog/introd...
@marimo.ioDay 4 of our launch week. One more announcement lands tomorrow.
Catch up on everything here: marimo.io/blog/launch...
@marimo.ioSee one quadratic programming notebook become a lecture, an explainer, and a hands-on lab. The athletes notebook becomes an overview, linked-chart explorer, and 3D field view.
Read @peter_gyarmati's announcement: marimo.io/blog/introd...
@marimo.ioShip for the audience you have: pre-render chosen interactive states as static files, with no Python runtime in production; run Python in the browser with Pyodide; or serve a live Python app with `marimo run`.
@marimo.ioStudio gives you two layers: pick the notebook outputs and values each audience sees, then design the view with HTML, React, Svelte, or any frontend stack. Each view lives beside the notebook, with results wired to marimo cells.
@marimo.ioIntroducing marimo-studio 🎬 One notebook, infinite views. Turn an analysis into a report, explorer, or 3D experience, all using the web framework you choose. Every result stays backed by the reactive Python cells.
@marimo.ioToday is day 3 of our launch week with even more exciting announcements coming tomorrow and on Friday!
Can you guess what’s coming next? See in our blog: marimo.io/blog/launch....
@marimo.ioYour agent can point back, too. Ask it to walk you through an analysis. Lens takes you from one output to the next, with a short explanation at each step. You move on when you’re ready. See how the results fit together and decide what to investigate next.
@marimo.io"Make this orange" "Why does this spike?" Just mark an output and leave a note. Your agent gets your selection, a screenshot, and the cell that produced it, plus its dependencies. Through marimo Pair, it can inspect the data, edit the code, and rerun the analysis.
@marimo.ioIntroducing marimo-lens 🔍
Point at a chart, mark a region, and tell your agent what to investigate or change.
Lens connects what you see to the code and data behind it. Your agent can point back, too.
Read blog, by @peter_gyarmati: marimo.io/blog/introd...
@marimo.ioRemote file storage is the second of five new features we're announcing this week, as part of our launch week!
Each day this week, we'll share something fun that the team has been cooking up.
We are only starting up, so follow along on what's next here: marimo.io/blog/launch....
@marimo.ioIf you are training a model, @huggingface can become your go-to remote connection for model and data storage.
And combining this with molab's RTX Pro 6000 GPUs it gives you everything you need to experiment with model training.
See example here: molab.marimo.io/notebooks/n...
@marimo.ioIf you are teaching, @github filesystem support allows you to provide assets for your students and read them directly from the notebook. Which means you can finally retire these `git clone` instructions from the class materials and reduce that friction.
