> ## Documentation Index
> Fetch the complete documentation index at: https://vastai-80aa3a82-docs-screenshot-updates.mintlify.site/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>

## 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).