> ## 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.

# Product Home

> Overview of CoreWeave compute, networking, and storage products for AI and HPC workloads

CoreWeave provides a cloud infrastructure platform optimized for AI training, inference, and high-performance computing (HPC) workloads. This page gives an overview of the products available to build, connect, and run your workloads.

### Compute

CoreWeave offers compute options for running containerized AI and HPC workloads:

<Columns cols={3}>
  <Card title="CoreWeave Kubernetes Service (CKS)" href="/products/cks" arrow="true">
    Managed Kubernetes on bare metal for training, inference, and HPC.
  </Card>

  <Card title="SUNK" href="/products/sunk" arrow="true">
    Slurm on Kubernetes for batch and burst workloads.
  </Card>

  <Card title="Inference" href="/products/inference" arrow="true">
    Serverless and dedicated inference options for serving AI models.
  </Card>

  <Card title="CoreWeave Sandbox" href="/products/sandboxes" arrow="true">
    Ephemeral compute sandboxes for agents and code interpretation.
  </Card>
</Columns>

### Storage

CoreWeave storage solutions support the data requirements of AI and ML workloads:

<Columns cols={2}>
  <Card title="CoreWeave AI Object Storage" href="/products/storage/object-storage" arrow="true">
    S3-compatible storage for datasets, model weights, and checkpoints.
  </Card>

  <Card title="Dedicated VAST Storage" href="/products/storage/dedicated-vast" arrow="true">
    Single-tenant VAST clusters with multi-protocol access at petabyte scale.
  </Card>

  <Card title="Distributed File Storage" href="/products/storage/distributed-file-storage/about" arrow="true">
    POSIX shared filesystem for multi-node access and distributed training.
  </Card>

  <Card title="Local Storage" href="/products/storage/local-storage" arrow="true">
    Ephemeral storage on GPU nodes for scratch space and caching.
  </Card>
</Columns>

### Networking

CoreWeave networking products create secure, high-performance connections between your resources and services:

<Columns cols={3}>
  <Card title="Virtual Private Clouds (VPCs)" href="/products/networking/vpc/about-vpcs" arrow="true">
    Isolated networks for CKS clusters.
  </Card>

  <Card title="HPC Interconnect" href="/products/networking/hpc-interconnect/about-hpc-interconnect" arrow="true">
    GPUDirect RDMA with InfiniBand for GPU-to-GPU communication.
  </Card>

  <Card title="Direct Connect" href="/products/networking/direct-connect/about-direct-connect" arrow="true">
    Private links through Equinix and Megaport.
  </Card>

  <Card title="IP addresses" href="/products/networking/ip/public-ipv4" arrow="true">
    Public IPv4 and Bring Your Own IP.
  </Card>

  <Card title="Ingress Service" href="/products/networking/ingress-service/create-public-dns-name" arrow="true">
    Public DNS names for services.
  </Card>
</Columns>