Private PyPI Server: How to Install Python Packages from a Custom Repository

Every Python developer knows how important package management is. If you have ever programmed in Python, you have probably used pip — one of the most popular tools for managing Python packages. But did you know that you can install packages from your own private PyPI server? In this article, I will explain what a Private PyPI Server is, what problems it solves, and how to use it from the terminal, in Jupyter Notebook, and with the Jupyter Package Manager plugin.

What is a Private PyPI Server

A private PyPI server is your own Python package repository. It works like the official PyPI, but instead of using the public service, the packages are stored on your own server. This gives you full control over which packages and versions can be installed. For this reason, private PyPI servers are often used by companies that want to manage dependencies in a safe and controlled way.

By using a private PyPI server, you can limit access to trusted packages only and reduce security risks.

Using a Private PyPI Server from the Terminal

Temporary usage

The simplest way is to pass the URL directly using the --index-url flag.

Command

pip install my_package --index-url https://my-pypi.example.com/simple

Example

Permanent configuration

You can also set a custom PyPI URL once, and pip will remember it. The easiest way is to set the index URL for the current user.

Command

pip config set global.index-url https://my-pypi.example.com/simple

pip config list # check configuration

pip install my_package # install package without using the flag

Example

Using a Private PyPI Server in Jupyter Notebook

Jupyter uses the same Python environment as pip. If you already configured pip in your terminal, you can install packages directly from a notebook and pip will automatically use your private PyPI server.

Code

import sys
!{sys.executable} -m pip install my_package

Example

Temporary usage

If you don't have a permanent configuration, or you don't want to use one, you can pass the URL using a flag, just like in the terminal.

This is also useful when you want to share a notebook with others who do not have a private PyPI configuration.

Code

import sys
!{sys.executable} -m pip install my_package --index-url https://my-pypi.example.com/simple

Example

Using a Private PyPI Server in Jupyter Package Manager

If you don't want to interfere with Python environments on your computer and want to save time on package management, you should try Jupyter Package Manager. It's an open-source extension for JupyterLab designed to simplify package management.

The extension allows you to install, update, remove, and list Python packages using a simple user interface. You do not need to write any code or run terminal commands, which makes package management faster and easier.

Installation

You can install the extension using following command:

pip install jupyter-package-manager

or download it from GitHub: https://github.com/mljar/package-manager

Want to learn more about the extension? Check out this article: Jupyter Package Manager

Open Jupyter Package Manager

After installation, a new icon should appear on the sidebar:

Go to settings

Click the settings icon in the top-right corner to open Package Manager Settings.

Change PyPI Server URL

To change the PyPI server URL, click Edit in the PyPI Server section:

Enter your custom PyPI URL and click Save:

That’s it — your PyPI server has been changed! If you want to switch back to the official PyPI, just click Restore defaults.

Go to Install Package

To verify that the new PyPI server works correctly, go back and click the package icon in the top-right corner (next to the settings icon). This will open the Install Package modal.

Install Package

Now enter the package name and click the Install button:

This extension and many others are available in MLJAR Studio - powerful Data Science tool that saves you time while coding with the help of an AI assistant and other features.

Get more information here: https://mljar.com/

Conclusion

Using a private PyPI server is a reliable way to manage Python packages in a secure and controlled environment. It is especially useful for companies and teams that need full control over dependencies and package versions. Remember to use Jupyter Package Manager when working in JupyterLab to save time and optimize daily work by making package management faster and easier.