Where do I find the editor password?

It is generated at install time and shown in your client area, on the application page. On the VPS it is also in the stack's .env file, under the ADMIN_PASSWORD key.

Where are my notebooks stored?

In the marimo-data Docker volume, mounted on /app inside the container. They are plain .py files: back up the volume or, better, version the folder in a Git repository.

Can I add other Python libraries?

The installed image already provides pandas, polars, numpy and altair. For other packages, the simplest route is to build an image derived from the official marimo image with your dependencies, then swap the image in the stack's compose file.

Kick off a software project quickly.

marimo

marimo — reactive Python notebooks on your own VPS: Git-friendly .py files, cells that recompute themselves, data and charts.

2 GB recommended RAM 1 vCPU (2 advised) Port 2718 In validation

In validation

Automatic installation for this solution is ready and currently going through our tests on a real server. Ordering will open as soon as validation is complete.

Source code

Tech stack

DockerPythonmarimo

Minimum RAM2 GB recommended

Minimum CPU1 vCPU (2 advised)

Default port2718

Compatible OSChoice of Linux distributions

marimo is an open-source Python notebook (Apache-2.0 licensed) designed as a successor to Jupyter. Every notebook is a plain .py file: it lives in Git, reads well in code review and runs as a script. Execution is reactive: when you change a cell, marimo automatically re-runs the cells that depend on it, so there is never hidden state or a stale result on screen.

Installed on your VPS, marimo becomes your team's data workbench: a password-protected web editor, notebooks that stay on your server, and the compute power you choose. The installed image already ships pandas, polars, numpy and altair to explore CSV or Parquet files, query an API and draw charts without installing anything.

A marimo notebook can also become a small web app: UI elements (sliders, dropdowns, tables) drive the computations, and `marimo run` serves the notebook read-only with the code hidden. Existing Jupyter notebooks convert with `marimo convert`.

Key features

Reactive execution — editing a cell re-runs the cells that depend on it, with no hidden state.

Pure Python notebooks — readable .py files, versioned in Git and runnable as scripts.

Data libraries preinstalled — pandas, polars, numpy and altair ready from the first start.

Built-in UI elements — sliders, dropdowns, tables and interactive charts wired to your code.

Notebook to app — marimo run publishes a notebook as a web app with the code hidden.

Migration from Jupyter — marimo convert turns an .ipynb file into a marimo notebook.

Password-protected editor — password generated at install time, shown in your client area.

When to use this solution?

1

Your team's data workbench

Explore CSV or Parquet exports, clean them with pandas or polars and share the notebook: colleagues review a .py file, not unreadable JSON.

2

Internal dashboards in Python

A notebook with sliders and dropdowns becomes an internal tool: the sales team filters the numbers without touching the code.

3

Training and teaching

Students and interns work on the same server, in the browser, without installing Python on their laptop; reactive execution makes every step of the reasoning visible.

4

Prototyping with your APIs and models

Call your internal APIs or a model served by Ollama on the same VPS, compare results and keep the prototype versioned in Git.

Deploy marimo on your VPS

Guide optimized for ServOrbit Cloud VPS.

01

Create the VPS

Order an Ubuntu 24.04 VPS with 2 GB of RAM: data libraries loaded in memory need it. A domain is optional — without one, the editor stays reachable through an SSH tunnel.

02

Install marimo from the Marketplace

In your client area, select marimo and click Install. The script deploys the official container, generates the editor password and creates the volume that keeps your notebooks.

03

Open the editor

With a domain attached (a subdomain of one of your domains, or the VPS's free subdomain marimo.{dns_slug}.servorbit-dns.com), the editor opens on https://<domain>, HTTPS included. Without a domain, open a tunnel: ssh -L 2718:127.0.0.1:<port> root@<vps-ip> then http://localhost:2718 — use the port shown in your client area (2718 by default).

04

Log in for the first time

marimo shows a login page asking for a password. It was generated at install time and is listed in your client area, on the application page. Enter it: the session stays open in that browser.

05

Create your first notebook

The editor opens the notebooks folder, which already contains the official tutorials. Create a new file, import pandas or polars and load a file: every edited cell automatically re-runs the cells that depend on it.

Frequently asked questions

Jupyter stores a notebook as JSON and runs cells in whatever order you launch them, which leaves stale results on screen. marimo stores a .py file and automatically re-runs the cells that depend on the one you edit: what you see always matches the code. Jupyter notebooks convert with marimo convert.

No. Without a domain, the editor is reachable through an SSH tunnel (ssh -L 2718:127.0.0.1:<port> root@<vps-ip>, then http://localhost:2718). To reach it directly over HTTPS, attach a subdomain of one of your domains or the VPS's free subdomain — nothing to buy.

Complementary solutions

Development

marimo isn't orderable yet.