> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/open-webui/open-webui/llms.txt
> Use this file to discover all available pages before exploring further.

# Quick Start

> Get started with Open WebUI from installation to your first chat in minutes

# Quick Start Guide

This guide will walk you through getting Open WebUI up and running quickly, from installation to your first conversation with an AI model.

## Prerequisites

Before you begin, ensure you have one of the following:

* Docker installed (recommended)
* Python 3.11 or 3.12 for pip installation
* A running Ollama instance (optional, for local models)

***

## Step 1: Choose Your Installation Method

<Tabs>
  <Tab title="Docker (Recommended)">
    The fastest way to get started is with Docker and the bundled Ollama image:

    <CodeGroup>
      ```bash With GPU theme={null}
      docker run -d -p 3000:8080 \
        --gpus=all \
        -v ollama:/root/.ollama \
        -v open-webui:/app/backend/data \
        --name open-webui \
        --restart always \
        ghcr.io/open-webui/open-webui:ollama
      ```

      ```bash CPU Only theme={null}
      docker run -d -p 3000:8080 \
        -v ollama:/root/.ollama \
        -v open-webui:/app/backend/data \
        --name open-webui \
        --restart always \
        ghcr.io/open-webui/open-webui:ollama
      ```
    </CodeGroup>

    Wait for the container to start, then navigate to [http://localhost:3000](http://localhost:3000)
  </Tab>

  <Tab title="Docker Compose">
    Create a `docker-compose.yaml` file:

    ```yaml theme={null}
    services:
      ollama:
        volumes:
          - ollama:/root/.ollama
        container_name: ollama
        pull_policy: always
        tty: true
        restart: unless-stopped
        image: ollama/ollama:latest

      open-webui:
        image: ghcr.io/open-webui/open-webui:main
        container_name: open-webui
        volumes:
          - open-webui:/app/backend/data
        depends_on:
          - ollama
        ports:
          - 3000:8080
        environment:
          - 'OLLAMA_BASE_URL=http://ollama:11434'
        extra_hosts:
          - host.docker.internal:host-gateway
        restart: unless-stopped

    volumes:
      ollama: {}
      open-webui: {}
    ```

    Start the services:

    ```bash theme={null}
    docker compose up -d
    ```

    Navigate to [http://localhost:3000](http://localhost:3000)
  </Tab>

  <Tab title="Python pip">
    Install Open WebUI using pip:

    ```bash theme={null}
    pip install open-webui
    ```

    Start the server:

    ```bash theme={null}
    open-webui serve
    ```

    Navigate to [http://localhost:8080](http://localhost:8080)
  </Tab>
</Tabs>

***

## Step 2: Create Your Admin Account

<Steps>
  <Step title="Access the Web Interface">
    Open your browser and navigate to:

    * **Docker**: [http://localhost:3000](http://localhost:3000)
    * **pip**: [http://localhost:8080](http://localhost:8080)
  </Step>

  <Step title="Sign Up">
    On first launch, you'll see the signup page. Create your admin account:

    * **Name**: Your display name
    * **Email**: Your email address (used for login)
    * **Password**: A secure password

    <Note>
      The first user to sign up automatically becomes the administrator.
    </Note>
  </Step>

  <Step title="Sign In">
    After creating your account, sign in with your credentials.
  </Step>
</Steps>

***

## Step 3: Set Up Your First Model

<Tabs>
  <Tab title="Using Ollama">
    If you're using the bundled Ollama image or have Ollama running, you can pull models directly from the UI.

    <Steps>
      <Step title="Access Settings">
        Click on your profile icon in the top-right corner and select **Settings**.
      </Step>

      <Step title="Navigate to Models">
        Go to the **Admin Panel** → **Settings** → **Models** section.
      </Step>

      <Step title="Pull a Model">
        In the model pull section, enter a model name (e.g., `llama3.2` or `mistral`) and click **Pull Model**.

        Popular models to start with:

        * `llama3.2` - Meta's latest Llama model
        * `mistral` - Mistral AI's efficient model
        * `phi3` - Microsoft's compact model
        * `gemma2` - Google's Gemma model
      </Step>

      <Step title="Wait for Download">
        The model will download and become available once complete. This may take a few minutes depending on the model size and your internet connection.
      </Step>
    </Steps>
  </Tab>

  <Tab title="Using OpenAI API">
    If you're using OpenAI or another compatible API:

    <Steps>
      <Step title="Access Settings">
        Click on your profile icon and select **Settings**.
      </Step>

      <Step title="Add API Connection">
        Go to **Admin Panel** → **Settings** → **Connections**.
      </Step>

      <Step title="Configure OpenAI">
        Enable the OpenAI API connection and enter:

        * **API Key**: Your OpenAI API key
        * **API Base URL**: `https://api.openai.com/v1` (or your custom endpoint)
      </Step>

      <Step title="Save Settings">
        Click **Save** to apply the configuration.
      </Step>
    </Steps>

    <Tip>
      You can also set the API key via environment variable:

      ```bash theme={null}
      -e OPENAI_API_KEY=your_secret_key
      ```
    </Tip>
  </Tab>
</Tabs>

***

## Step 4: Start Your First Chat

<Steps>
  <Step title="Create a New Chat">
    Click the **+ New Chat** button on the sidebar or in the center of the screen.
  </Step>

  <Step title="Select a Model">
    Click on the model selector dropdown at the top of the chat interface and choose your preferred model.
  </Step>

  <Step title="Send Your First Message">
    Type a message in the text box at the bottom and press Enter or click the send button.

    Example prompts to try:

    * "Explain quantum computing in simple terms"
    * "Write a Python function to reverse a string"
    * "What are the best practices for Docker deployments?"
  </Step>

  <Step title="Explore Features">
    Try these features in your chat:

    * **Multi-line input**: Press Shift+Enter for new lines
    * **Regenerate**: Click the regenerate button to get a different response
    * **Copy**: Copy responses with the copy button
    * **Voice input**: Click the microphone icon for voice-to-text
  </Step>
</Steps>

***

## Step 5: Explore Advanced Features

Now that you have Open WebUI running, explore these powerful features:

<CardGroup cols={2}>
  <Card title="Document Upload" icon="file">
    Upload documents and ask questions about their content using the RAG feature. Click the **+** icon in the chat input to upload files.
  </Card>

  <Card title="Web Search" icon="globe">
    Enable web search to get real-time information. Type `#` followed by a search query or URL.
  </Card>

  <Card title="Multiple Models" icon="layer-group">
    Compare responses from multiple models simultaneously by selecting multiple models from the dropdown.
  </Card>

  <Card title="Custom Prompts" icon="pen">
    Create and save custom prompts for repeated tasks in the Workspace section.
  </Card>
</CardGroup>

***

## Common Tasks

### Upload and Query Documents

<Steps>
  <Step title="Enable RAG">
    RAG (Retrieval Augmented Generation) is enabled by default.
  </Step>

  <Step title="Upload Document">
    Click the **+** icon next to the chat input and select **Upload Files**. Choose your document (PDF, TXT, DOCX, etc.).
  </Step>

  <Step title="Ask Questions">
    Once uploaded, ask questions about the document:

    * "Summarize this document"
    * "What are the key points in section 3?"
    * "Extract the main findings"
  </Step>
</Steps>

### Create a Custom Model

<Steps>
  <Step title="Access Workspace">
    Click **Workspace** in the sidebar, then navigate to **Models**.
  </Step>

  <Step title="Create Model">
    Click **+ Create Model** and configure:

    * **Base Model**: Select the foundation model
    * **System Prompt**: Define the model's behavior
    * **Parameters**: Adjust temperature, top\_p, etc.
  </Step>

  <Step title="Save and Use">
    Save your custom model and select it from the model dropdown in any chat.
  </Step>
</Steps>

### Enable Web Search

<Steps>
  <Step title="Configure Search Provider">
    Go to **Admin Panel** → **Settings** → **Web Search**.
  </Step>

  <Step title="Select Provider">
    Choose a search provider (SearXNG, Google PSE, Brave Search, DuckDuckGo, etc.) and configure any required API keys.
  </Step>

  <Step title="Use in Chat">
    In your chat, type `#` followed by a search query to inject web results into the conversation.
  </Step>
</Steps>

***

## Configuration Quick Reference

### Environment Variables

Set these via Docker `-e` flag or in your environment:

```bash theme={null}
# Ollama connection
OLLAMA_BASE_URL=http://localhost:11434

# OpenAI API
OPENAI_API_KEY=your_secret_key
OPENAI_API_BASE_URL=https://api.openai.com/v1

# Security
WEBUI_SECRET_KEY=your_secret_key

# Disable signup after initial setup
ENABLE_SIGNUP=False
```

### Admin Settings Location

Access admin settings at:

1. Click your **profile icon** (top-right)
2. Select **Admin Panel**
3. Navigate to **Settings**

Key sections:

* **General**: Basic configuration
* **Models**: Model management and settings
* **Connections**: External API connections
* **Users**: User management and permissions

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="Models not appearing">
    **Solution**:

    1. Verify Ollama is running: `docker ps | grep ollama`
    2. Check the `OLLAMA_BASE_URL` environment variable
    3. Ensure models are pulled: `docker exec -it ollama ollama list`
  </Accordion>

  <Accordion title="Cannot connect to Ollama">
    **Solution**:

    1. If using separate containers, ensure they're on the same network
    2. Use `--add-host=host.docker.internal:host-gateway` for local Ollama
    3. For bundled installation, no additional configuration needed
  </Accordion>

  <Accordion title="Slow model responses">
    **Solution**:

    1. Use smaller models (phi3, mistral) for faster responses
    2. Enable GPU support with `--gpus all` and `:cuda` image
    3. Adjust model parameters (lower temperature, max\_tokens)
  </Accordion>

  <Accordion title="Login issues">
    **Solution**:

    1. Clear browser cache and cookies
    2. Verify `WEBUI_SECRET_KEY` is set and consistent
    3. Check browser console for errors (F12)
  </Accordion>
</AccordionGroup>

***

## Next Steps

<CardGroup cols={2}>
  <Card title="Configuration Guide" icon="gear" href="/configuration">
    Learn about all configuration options and environment variables
  </Card>

  <Card title="Features Overview" icon="sparkles" href="/features">
    Explore all features including RAG, web search, and image generation
  </Card>

  <Card title="User Management" icon="users" href="/admin/users">
    Set up user roles, permissions, and access control
  </Card>

  <Card title="API Reference" icon="code" href="/api">
    Integrate Open WebUI with your applications
  </Card>
</CardGroup>

***

## Tips for Success

<Tip>
  **Disable Signup After Setup**: Once you've created accounts for your users, disable signup to prevent unauthorized access:

  ```bash theme={null}
  -e ENABLE_SIGNUP=False
  ```
</Tip>

<Tip>
  **Use Persistent Storage**: Always use volume mounts to prevent data loss:

  ```bash theme={null}
  -v open-webui:/app/backend/data
  ```
</Tip>

<Tip>
  **Start Small**: Begin with smaller models like `phi3` or `mistral` to test the setup before downloading larger models.
</Tip>

Need help? Join our [Discord community](https://discord.gg/5rJgQTnV4s) or check the [full documentation](https://docs.openwebui.com/).