There are many ways to install Python. The official site, system package managers, Anaconda, pyenv — they all work, but they all leave you wiring tools together yourself.

Instead, we’ll use uv, a modern tool that installs Python, manages versions, creates virtual environments, and installs packages. One tool for the whole job. It’s fast — usually 10 to 100 times faster than the old pip workflow — and it’s quickly becoming the standard in the Python and AI world.

Install uv

macOS and Linux

Open a terminal and run:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows

Open PowerShell and run:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Verify uv is installed

Close and reopen your terminal, then run:

uv --version

You should see something like:

uv 0.5.0

The exact version will be different — what matters is that the command runs.

Install Python using uv

With uv installed, getting Python is one command:

uv python install 3.13

This downloads Python 3.13 (or a newer version, depending on what’s current) and stores it in a place uv manages. You don’t need to think about where.

Check that Python is available:

uv run python --version
Python 3.13.0

The uv run part tells uv to run a command using its managed Python. We’ll be using uv run for the rest of the course.

Why not just install Python from python.org?

You can. The installer at python.org works fine, and millions of people use it. The reason we don’t:

  • Multiple Python versions get messy. One project needs 3.11, another needs 3.13. Tools like uv keep them straight without your help.
  • Virtual environments are tedious. Without uv, you have to run python -m venv .venv then activate it then remember to install packages inside it. uv does it all in one step.
  • pip is slow. On a machine learning project with hundreds of dependencies, pip install can take minutes. uv does it in seconds.

If you already have Python installed from somewhere else, that’s fine — leave it alone. uv will use its own copy.

A note on the python command

Older tutorials say “run python script.py”. On many systems that command doesn’t exist or points to the wrong version. Throughout this course we’ll use:

uv run python script.py

This always runs Python through uv, using whichever version uv has installed. You’ll see the shorter form python in other docs — they mean the same thing once your environment is set up.

What’s next

Python is installed. Next we’ll meet the REPL — Python’s interactive prompt, where you can try ideas one line at a time.

Toggle theme (T)