> ## Documentation Index
> Fetch the complete documentation index at: https://www.latitude.sh/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Operating systems

Latitude.sh provides a variety of Operating systems that you can use to create your servers, such as Windows and Linux distributions. You can find all supported operating systems by going to the [server create page](https://metal.new) or through the following API request.

```shell theme={null}
curl --request GET \
     --url https://api.latitude.sh/plans/operating_systems \
     --header 'Authorization: Bearer <API_KEY>'
```

## ML-in-a-Box

ML-in-a-Box is a custom image built by Latitude.sh that you can use to get started quickly with any of the GPU servers available. This image comes with several dependencies needed for Machine Learning and AI workloads and is automatically selected when you are deploying a GPU server and select Ubuntu as the Operating System.

The versions of the installed dependencies are optimized for compatibility and might not be the latest versions available.

| Depdendency              | Version     |
| :----------------------- | :---------- |
| Python                   | 3.10        |
| NVIDIA CUDA              | 13.0        |
| NVIDIA Driver            | 580.105.08  |
| NVIDIA Container Toolkit | 1.18.0-1    |
| Docker                   | 29.0.0      |
| PyTorch                  | 2.9.0+cu130 |

## Custom images

Installing your own images is supported through iPXE scripts. Go to [Custom image (iPXE)](/servers/custom-images) to learn more.