---
title: Quickstart
description: Cortex Quickstart.
keywords:
  [
    Jan,
    Customizable Intelligence, LLM,
    local AI,
    privacy focus,
    free and open source,
    private and offline,
    conversational AI,
    no-subscription fee,
    large language models,
    Cortex,
    Jan,
    LLMs
  ]
---

import { Callout, Steps } from 'nextra/components'
import { Cards, Card } from 'nextra/components'

# Quickstart

<Callout type="warning">
🚧 Cortex is under construction.
</Callout>

To get started,  confirm that your system meets the [hardware requirements](/cortex/hardware), and follow the steps below:

```bash
# 1. Install Cortex using NPM
npm i -g @janhq/cortex

# 2. Download a GGUF model
cortex models pull llama3

# 3. Run the model to start chatting
cortex models run llama3

# 4. (Optional) Run Cortex in OpenAI-compatible server mode
cortex serve
```
<Callout type="info">
For more details regarding the Cortex server mode, please see here:
- [Server Endpoint](/cortex/server)
- [`cortex serve` command](/cortex/cli/serve)
</Callout>

## What's Next?
With Cortex now fully operational, you're ready to delve deeper:
- Explore how to [install Cortex](/cortex/installation) across various hardware environments.
- Familiarize yourself with the comprehensive set of [Cortex CLI commands](/cortex/cli) available for use.
- Gain insights into the system’s design by examining the [architecture](/cortex/architecture) of Cortex.