This page explains how to install the Sabana CLI and SDK

CLI (v0.1.8)

Download the CLI by copying and running the following command in a terminal shell.

M1 (Arm64)

curl --output && unzip && rm

Intel (x86_64)

curl --output && unzip && rm


curl --output sabana.tar.gz && tar xf sabana.tar.gz && rm sabana.tar.gz


curl --output sabana.tar.gz && tar xf sabana.tar.gz && rm sabana.tar.gz
We currently support Windows via Windows Subsystem for Linux​
The binary now is ready to use. You could move sabana to a more convenient location part of your PATH environment variable. One way to check that the CLI is working properly is by running:
./sabana help

Log in

Once your account is created and you have credentials, then you can login with:
sabana login
This will display a web url that you can open in a browser for entering your email and password. You can check that your credentials are correctly configured by running the following command:
sabana ls
This is the expected output:
| Image | Tags |
You can also list the images built by our robot user and check the pre-built images available:
sabana ls robot

About credentials

Currently, credentials are valid for up to 30 days after you login. This means that if you find issues logging in at any point in time, then try to login again and generate new tokens. Additionally, after a successful login, there will be a file created at your home folder ~/.sabana/config.json containing the tokens for accessing our services and APIs.
Don't make the config.json file public in any way, because it would give others access to your account
In some cases, you may want to use your credentials in a headless server (without any graphical interface). For this case we recommend login in locally in your computer first and then copying over your sabana directory, for example:
scp -r ~/.sabana [email protected]:
Lastly, you can remove credentials from the current system by doing:
sabana logout

SDK (v0.1.23)

Currently our SDK is based in Python and distributed as a pip package. Use the pip command to install the package in your Python environment. We currently support Python versions above 3.8
python3 -m pip install sabana==0.1.23
You can use the following snippet to test that the client has been successfully installed
python3 -c "import sabana; print(sabana.__version__)"
If it runs without exceptions then you have successfully installed our Python SDK. You are now ready to build, deploy, and run hardware in the cloud.