Turn Jupyter Notebook to Web App

Python Notebooks
to Web Apps

Turn your Python Notebook into a Web App with the open-source Mercury framework. Share your results with non-technical users.

Sign up for Cloud

Check our GitHub

What you can build?

Web Apps

Create data-rich web apps with interactive widgets

Reports

Build reports with automatic schedule and PDF export

Dashboards

Visualize your results in interactive dashboards

Websites

Easily turn notebooks into beautiful static websites

Presentations

Interactive slides that are recomputed during the show

REST API

Execute notebooks with REST API and return JSON response

"After changing to use Mercury framework, I get to simplify my code a lot without having to deal with the widget listeners and layout "

Kuanmy
Software Developer

"Mercury provides useful widgets so my apps are interactive, but with desired limits set since users can not edit the code. The layout provided by Mercury framework makes anyone feel comfortable because of its clean and flawless look. Knowing this framework made my collaborations much more efficient. "

Katarzyna-Anna Bensch
Data Analyst

"Mercury seems to be very interesting. Just had a try today and was able to fire up a GUI within 10 minutes that would have taken me at least 2h using Panel. Really nice thing. "

Hugo Santos Silva
Data Engineer

Web Apps created with Mercury

Check example web applications created in Python notebook with Mercury

More examples
Example static notebook

Share notebooks as static websites

Live appCheck code
Python Notebook with financial data application

Ticker report created in Python notebook

Live appCheck code
Python Notebook with dashboard created with altair viz

Interactive dashboard built with Altair

Live appCheck code

One click deployment

Sharing notebooks with Mercury is very easy.
The easiest way is to use Mercury Cloud or you can setup your own server.

Mercury Cloud

Mercury Cloud is a service managed by us. You can setup your own website with unique domain address in few seconds. Notebooks deployment is done with one click.

Create website

Self host

You can configure your own server and deploy notebooks on your own infrastructure. We provide instructions on how to run your own server with docker-compose.

Documentation

Build web apps with Python only

You don't need to know frontend frameworks, HTML, CSS or JavaScript.

Create web applications in pure Python without leaving Jupyter Notebook.

Documentation

How does it work?

No callbacks!

Mercury automatically re-executes cells below updated widget. You don't need to write callbacks to handle widget update.

Perfect layout

Widgets are in the sidebar and outputs will appear in the same order as in the notebook.

How does it work?

1

Start with Python Notebook

Python Notebook is a great tool to connect Markdown, Python code and outputs into meaningful documents.

However, sharing notebooks with others might be challenging!

Python Notebook

2

Make it interactive

Add widgets to you Python Notebook and make it interactive.

Install Mercury framework

pip install mercury

Import Mercury and use widgets during development

import mercury as mr
text_input = mr.Text(label="Your name?")

Check available widgets in docs

Input Widgets

Output Widgets

Add widgets

3

Run as Web App

Start Mercury Server and serve your notebook as Web App

mercury run

Users can execute your code but they can't edit it!

Mercury App

Amazing features!

Hide or show code

You decide if you want to show or hide code in the app

Export PDF / HTML

Easily export the executed notebook as a PDF or HTML

Authentication

Decide with whom to share your notebooks

Multiple apps

You can share as many notebooks as you want

Embedding

Run notebook with Mercury and embed it on any website

Output files

Produce files in notebook and make them downloadable

Works great with

Matplotlib
Plotly
Altair Viz
IpyVizzu
Seaborn
Pandas
Numpy
Scikit-Learn
OpenCv
Tensorflow
PyTorch
Sympy

Basically every Python package!

Yes, it is open-source!

Mercury is an open-source framework. It is available on AGPL v3 license. We offer a commercial-friendly license with more features, dedicated support and private forks.

Star on GitHub