> ## 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.

# Workloads

> Overview of SUNK and CoreWeave Inference for running AI workloads on CoreWeave

# Training and inference for AI workloads

Workloads brings together SUNK for running Slurm on Kubernetes and CoreWeave Inference for serving models on NVIDIA GPUs.

Choose a product to get started.

[SUNK
Slurm on Kubernetes for batch and burst workloads.](https://docs.coreweave.com/products/sunk.md)

[CoreWeave Inference
Serverless and dedicated inference options for serving AI models.](https://docs.coreweave.com/products/inference.md)

## Run agent workloads in Sandboxes

[CoreWeave Sandboxes](https://docs.coreweave.com/products/sandboxes.md) run agent tools and evaluations on serverless capacity or your own CoreWeave Kubernetes Service (CKS) clusters. The shared documentation is under **Forge > Sandboxes**. Use a Forge API key or CoreWeave API access token for serverless capacity. Use a CoreWeave token for CKS.

Start with the [serverless quickstart](https://docs.coreweave.com/products/sandboxes/get-started.md), or [use your own compute](https://docs.coreweave.com/products/sandboxes/get-started-cks.md) to configure a CKS runner. The same [SDK guides](https://docs.coreweave.com/products/sandboxes/client.md), [GPU guide](https://docs.coreweave.com/products/sandboxes/gpu-sandboxes.md), and [computer-use tutorial](https://docs.coreweave.com/products/sandboxes/tutorials/computer-use.md) cover both placement modes.
