# Arrakis docs

Rent GPUs by the hour from the console, the CLI or your agent. Everything you need, from first sign-in to an agent with a spend cap.

## Start

- [Quickstart](/docs/quickstart.html.md): From sign-in to an SSH session on a rented GPU in about five minutes.

## Use Arrakis

- [Rent and connect](/docs/rent.html.md): Find the right machine, choose its size and image, connect over SSH, and stop it on time.
- [Deploy a model](/docs/models.html.md): Serve any Hugging Face model (text, image, video or speech) with vLLM, SGLang, vLLM-Omni or llama.cpp and your own engine options: an OpenAI-style endpoint with its own API key, and root SSH.
- [Images](/docs/images.html.md): What a machine boots: an Arrakis image with SSH and the usual tools, or an image of your own, public or private.
- [Volumes](/docs/volumes.html.md): Extra storage that outlives your machines: mounted at /mnt, billed on the data it holds, apart from the machine's own disk.
- [Billing](/docs/billing.html.md): Prepaid, by the hour, with a 1-hour minimum, and a breaker that stops machines before you owe anything.

## Build

- [Agents and MCP](/docs/agents.html.md): Give an agent a token it can't overspend, and connect it over MCP.
- [CLI](/docs/cli.html.md): Search, rent and manage machines from your terminal with the arrakis command.
- [API](/docs/api.html.md): Authenticate with a token, then search, rent, manage and pay over plain JSON.

## Trust

- [Security](/docs/security.html.md): How Arrakis protects your account, your money and your machines, and what stays your responsibility.

## Reference

- [OpenAPI](/openapi.json)
- [Interactive API](/docs/reference)
- [llms.txt](/llms.txt)