Nº27 · Visualization
Quarto
Publish reproducible reports, dashboards, and sites from your notebooks and code.
What is it?
Quarto is an open-source technical and scientific publishing system, built on
Pandoc. You write .qmd files that mix Markdown with code cells (Python, R,
Julia, Observable), and Quarto executes and renders them into a final document: HTML,
PDF, Word, presentations, dashboards, books, or a full website.
It is the successor to R Markdown, now language-agnostic: for the Python stack it runs the code through Jupyter kernels.
What is it used for?
- Reproducible reports where the narrative, the code, and its output live in the same file — re-render and the numbers and charts recompute themselves.
- One source, many formats: the same
.qmdexports to HTML, PDF, Word, slides (reveal.js), or a site, without rewriting the content. - Dashboards and technical documentation sites, versionable in git like any text file.
When to use it / when not?
Use it when the deliverable is a document or report that combines analysis and explanation, and you want it reproducible and exportable to several formats. It fits perfectly on top of a Jupyter notebook and charts made with Matplotlib or seaborn: write the analysis once and publish it wherever you need.
Think twice if what you need is an interactive BI dashboard with live SQL filters for business users — that is what Superset is built for. If you only want to explore data without publishing anything, plain Jupyter is enough. And if you're after an interactive data app, look at Streamlit or Dash.
Get started in 1 minute
# 1. Install the Quarto CLI (pick one):
# - Official installer: https://quarto.org/docs/get-started/
# - Homebrew (macOS): brew install quarto
# - pip (bundles the binary): pip install quarto-cli
# 2. Create a project and preview it with live reload
quarto create project default report
cd report
quarto preview
A minimal .qmd with a Python cell:
---
title: "My first report"
format: html
---
## Sales by month
```{python}
import matplotlib.pyplot as plt
plt.plot([1, 2, 3], [10, 40, 25])
plt.show()
```
# Render to other formats from the same file
quarto render report.qmd --to pdf # or docx, revealjs...
Quick trivia — test what you just read.
How much do you know about Quarto?
Official documentation
The source of truth lives there. Here we orient you; the depth is up to you.
Open official docs ↗What to learn next
See alsoKioskoAI mailbox
Anything missing from the Quarto guide?
This decides what gets written next. If it only half-helped, say so.
Nº27 · Updated 2026-07-23