---
title: "Quickstart"
description: "Run RAGNA Studio on your own machine for a local try-out."
---

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

# Quickstart

This runs the whole stack on localhost. It needs no proxy and no TLS. For a server setup, see [Self-hosting](/self-hosting).

## Before you start

You need Docker with Docker Compose, `git` and `make`. You also need:

- One S3-compatible bucket setup, see [Storage](/self-hosting/storage).
- One OAuth app (Google or Microsoft), see [Configuration](/self-hosting/configuration#oauth).
- At least one LLM API key.

## Run it

1. **Get the code**

   ```bash
   git clone https://github.com/ragna-ai/ragna-studio.git
   cd ragna-studio
   cp .env.example .env
   ```
2. **Fill in .env**

   Set the secrets, storage, one OAuth provider and one AI key. The [environment variables](/reference/environment-variables) reference lists every setting.
3. **Start the stack**

   ```bash
   make up-dev-full
   ```

   This starts the whole stack with all ports published.
4. **Open the app**

   Open [localhost:3000](http://localhost:3000). The API is on port 3010.

> **Not for public servers**
>
> The dev compose publishes every port, including Postgres and Redis. Do not use it on a public server.

Source: https://docs.ragna.io/getting-started/quickstart/index.mdx