jupytext

What it is and what it does

Jupytext bridges the gap between Jupyter notebooks and plain-text development workflows by allowing notebooks to be stored and edited as regular Python scripts, Markdown files, or other text formats. It reads `.ipynb` files and converts them to formats like `.py` (percent-format) or `.md` (Markdown-based), and vice versa, preserving cell structure and metadata. The tool integrates with JupyterLab as an extension, making text-format notebooks appear and behave like regular notebooks in the interface.

The package enables two main workflows: text-only notebooks (inputs only, no outputs) for clean version control and IDE editing, and paired notebooks (simultaneous `.ipynb` and `.py` or `.md` files) that preserve outputs while allowing text-based collaboration. It includes command-line tools for batch conversion, synchronization, and piping notebooks through linters. Dependencies are lightweight—markdown-it-py, mdit-py-plugins, nbformat, packaging, pyyaml, and tomli—and the package supports Python 3.9 through 3.14.

Use it for:

• Store Jupyter notebooks in Git with meaningful diffs by pairing `.ipynb` with `.py` or `.md` and committing only the text version.
• Edit notebook code in your IDE (VS Code, PyCharm, etc.) as a regular Python script, then reload it in Jupyter to see outputs.
• Collaborate on notebooks via Git pull requests, where each collaborator can review and merge text-format changes without output noise.
• Convert existing `.ipynb` notebooks to `.py` or `.md` format for archival, documentation, or integration into a code repository.
• Automate notebook-to-script conversion in CI/CD pipelines using the command-line interface.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Jupytext converts Jupyter notebooks to and from plain-text formats (Python scripts, Markdown, or other text files) while preserving notebook structure, enabling version control and IDE editing of notebook content.

Yes. Jupytext is actively maintained, has no known vulnerabilities, and solves a real problem for teams wanting to version-control notebooks and edit them in IDEs. Install it if you work with Jupyter and Git together, or need to treat notebooks as code. The low dependency count and permissive MIT license make it a safe addition to any Python environment.

Install

jupytext on PyPI

pip install jupytext

uv add jupytext

poetry add jupytext

Installing jupytext

Before you install

Low friction install with six straightforward dependencies (markdown-it-py, mdit-py-plugins, nbformat, packaging, pyyaml, tomli). Active maintenance with a recent release (24 days ago) and 7228 repository stars.

License in practice

MIT License permits unrestricted use, modification, and distribution with minimal restrictions—suitable for both open-source and commercial projects.

Quickstart

pip install jupytext

import jupytext
notebook = jupytext.read('notebook.ipynb')
jupytext.write(notebook, 'notebook.py')

Requires JupyterLab or Jupyter to be installed in the same Python environment; command-line usage also available without Jupyter.

Verify before relying

• Whether paired notebook sync handles merge conflicts automatically or requires manual intervention beyond the described pop-up prompts.
• Performance characteristics when working with very large notebooks or many paired files.
• Compatibility with notebook extensions other than the Jupyter Collaboration extension mentioned.

Package facts

Evidence: jupytext-1.19.5-py3-none-any.whl

Tags

More Markup packages