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

# Welcome

[Vast.ai](https://www.vast.ai) is a marketplace for affordable GPU cloud computing. We make it easy for anyone to:

* Spin up GPU instances in seconds at competitive [prices](/guides/instances/pricing).
* Scale across thousands of GPUs from [Secure Cloud datacenters](https://cloud.vast.ai/create/?instanceDiskSizeMin=16\&instanceDurationMin=259200\&instanceType=onDemand\&isAvxSupported=false\&isHostSecure=true\&isMachineIpStatic=false\&isOfferAvailable=true\&isOfferCompatible=true\&isOfferVerified=true\&isQueryInverted=false\&machineReliabilityMin=0.9\&sorts=score-desc) or community providers.
* Launch [prebuilt](/guides/templates/introduction) or [custom templates](/guides/templates/creating-templates) with one click.

## How It Works

Vast.ai connects compute providers, from hobbyists to Tier-4 datacenters, with users who need GPUs. Our [search engine](/guides/instances/choosing/find-and-rent) lets you filter by GPU type, RAM, CPU, bandwidth, and more, while providers retain full control over [pricing](/guides/instances/pricing) and [contracts](/guides/instances/choosing/instance-types).

## What are you looking for?

<CardGroup cols={2}>
  <Card title="Rent your first GPU" href="/guides/get-started/quickstart" icon="rocket">
    5-minute walkthrough: search the marketplace, launch an instance, SSH in.
  </Card>

  <Card title="Automate with the CLI or SDK" href="/cli/hello-world" icon="terminal">
    `vastai` from the shell or Python. Same operations, different syntax.
  </Card>

  <Card title="Pricing" href="/guides/pricing" icon="tag">
    On-demand, reserved, interruptible, and serverless rates.
  </Card>

  <Card title="Connect to instances" href="/guides/instances/connect/overview" icon="plug">
    SSH, Jupyter, or the web portal.
  </Card>

  <Card title="Build a template" href="/guides/templates/creating-templates" icon="layer-group">
    Package your environment so any GPU can run it with one click.
  </Card>

  <Card title="See example workloads" href="/pytorch" icon="cube">
    PyTorch, vLLM, ComfyUI, agents, and more. Real deployments you can copy.
  </Card>

  <Card title="Run serverless workloads" href="/guides/serverless/quickstart" icon="bolt">
    Deploy auto-scaling endpoints for inference, batch jobs, or APIs.
  </Card>

  <Card title="Host GPUs and earn" href="/host/hosting-overview" icon="server">
    List your machines on the marketplace, set prices, get paid.
  </Card>
</CardGroup>

## Mission

Vast.ai's mission is to align and democratize AI. Machine learning is progressing towards powerful AI systems with the potential to radically reshape our future. We believe it is imperative that this awesome power be distributed widely; that its benefits accrue to the many rather than the few; that its secrets are unlocked for the good of all humanity. Towards these ends we work to ensure that the compute powering AI is supplied by the people and for the people.

## Talk to Us

* **Support Chat** → Available 24/7 in the bottom-right corner of our [console](https://cloud.vast.ai).
* **Email** → [contact@vast.ai](mailto:contact@vast.ai)
* **Discord** → Join our [community](https://discord.gg/vast) for help and discussions.
* **Documentation** → Browse our [instances guide](/guides/instances/overview), [templates documentation](/guides/templates/introduction), or [troubleshooting tips](/guides/reference/troubleshooting).