---
title: "Chat"
description: "Use the cluster's models in your browser through Open WebUI."
---

> 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.

# Chat

Chat runs on [Open WebUI](https://docs.openwebui.com) at [chat.c4e.ai](https://chat.c4e.ai). After you [sign in](https://docs.c4e.ai/getting-started/sign-in/), you land on a new chat.

## Pick a model

The model menu at the top of a chat lists the models your account can use. Everyone has **DeepSeek V4.1 Flash**. In the API, the same model is `DeepSeek-V4.1-Flash-1017`. If an admin has given you more, they show up in the same menu.

Model names stay the same even when we change the engine that serves them, so a name you use today keeps working.

## When the models are paused

The models run as a job on the same Sparks as everyone's research jobs. While the cluster has only two Sparks, a research job pauses the models, and they come back by themselves when the job ends. While they're paused, chat answers with a message saying the models are paused and come back on their own. Try again in a few minutes. [Slurm jobs](https://docs.c4e.ai/getting-started/slurm-jobs/#jobs-pause-the-shared-models) explains why.

## Your data

Your chats are stored on the cluster so you can come back to them. Don't paste passwords, keys or anything you wouldn't want an admin of a shared machine to be able to see.

Source: https://docs.c4e.ai/getting-started/chat/index.mdx
