---
title: "C4E Cluster"
description: "A shared NVIDIA DGX Spark cluster for students. Chat with the cluster's models, call them from code, and run your own GPU jobs."
---

> Documentation Index
> Fetch the complete documentation index at: https://docs.c4e.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# C4E Cluster

C4E is a small research cluster of NVIDIA DGX Sparks. Two are online today and four more are coming. Everyone invited can chat with the cluster's models and call them from code, and people with job access can run their own GPU work through Slurm.

## What you can do

- **Chat** — Talk to the cluster's models in your browser at chat.c4e.ai.
- **Use the API** — Make an API key and call the models with the OpenAI SDK.
- **Run jobs** — Upload files and submit Slurm jobs from Open OnDemand.
- **Use with your agent** — Let Claude or another agent run your Slurm jobs through mcp.c4e.ai.

## Start here

1. **Sign in**

   Open [chat.c4e.ai](https://chat.c4e.ai) and sign in with the code we email you. See [Sign in](https://docs.c4e.ai/getting-started/sign-in/).
2. **Pick what you need**

   Chat in the browser, make an [API key](https://docs.c4e.ai/getting-started/api/), or open [Open OnDemand](https://docs.c4e.ai/getting-started/ondemand/) if you have job access.

The cluster is invite-only. If you can't sign in, or you need job access, email admin@c4e.ai.

To see how the pieces fit together, read [Architecture](https://docs.c4e.ai/architecture/).

Source: https://docs.c4e.ai/index.mdx
