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

# About instances

> Learn about CoreWeave compute instances and how to choose the right hardware for your workloads

CoreWeave offers a range of compute instances for cloud workloads. Instances are the physical machines that run on CoreWeave's infrastructure. Each instance provides dedicated GPUs, CPUs, memory, storage, and networking resources used by CKS Nodes, LLM inference workloads, and other high-performance computing (HPC) workloads.

<Columns cols={3}>
  <Card title="Selecting an instance" href="/platform/instances/selecting-an-instance" horizontal arrow="true">
    Key considerations when selecting the right instance for your LLM workload
  </Card>

  <Card title="About NVL72-powered instances" href="/platform/instances/nvl72" arrow="true">
    NVL72-powered instances deliver high performance, efficiency, and scalability
  </Card>

  <Card title="GPU instance details" href="/platform/instances/gpu-instances" arrow="true">
    Detailed specifications and use cases for GPU instances
  </Card>

  <Card title="CPU instance details" href="/platform/instances/cpu-instances" arrow="true">
    CPU instance options and performance characteristics
  </Card>

  <Card title="Instance pricing" href="https://www.coreweave.com/pricing" arrow="true">
    Pricing for all instance types
  </Card>

  <Card title="Regional availability" href="/platform/instances/availability-matrix" arrow="true">
    Regional availability for all instance types
  </Card>
</Columns>