Getting started with FastAPI on a Raspberry Pi

First published:

Last Edited:

Number of edits:

Recommended operating system: Ubuntu Server 20.4.

After loading the operating system on an SSD card, booting, and logging into it, the most important is to perform an update to the latest libraries following the steps that will appear on screen:

sudo apt full-update

apt-get is the package manager of Ubuntu, and the command apt is a convenient way of using it. sudo instructs the command-line to run the command as the "Super User", a user that can make changes to the entire computer. Use sudo with care.

In order to get started with Python, we must install some libraries:

sudo apt install python3 python3-pip python3-venv

This will install Python (most likely it is already installed, but just in case), the Python package manager pip, and the venv module to create virtual environments.

See: Difference conda environment and virtual environment in Pyhon

Once it is ready, we can proceed to creating a virtual environment, for example:

python3 -m venv fastapi

The -m venv is a quick way of using a Python module directly from the command line. To activate the virtual environment we must run:

source fastapi/bin/activate

To deactivate the virtual environment:


We can see that it worked because there'll be a (fastapi) string at the beginning of the command line.

Before installing FastAPI, we will need to install cargo, the package manager of Rust, another programming language.

sudo apt install cargo

Now we can install the python packages that are needed. It is important to install these packages inside a virtual environment, for example by first activating fastapi as we did earlier.

pip install fastapi[all]

It is important to note that all the commands that involve apt are acting at a computer level. Cargo is installed on the computer, not within the virtual environment, for example. Any changes to it will have impact on all the virtual environments that rely on it.

To start a FastAPI program and make it accessible through the network, from an outside computer or device:

uvicorn main:app --reload --host


These are the other notes that link to this one.

Nothing links here, how did you reach this page then?


Join my experiment of better thinking and interesting discussions
Aquiles Carattino
Aquiles Carattino
This note you are reading is part of my digital garden. Follow the links to learn more, and remember that these notes evolve over time. After all, this website is not a blog.
© 2021 Aquiles Carattino
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License
Privacy Policy