> ## 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.

# Frequently Asked Questions

> Answers to common questions about Open WebUI

## General Questions

<AccordionGroup>
  <Accordion title="What is Open WebUI?">
    Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform designed to operate entirely offline. It supports various LLM runners like Ollama and OpenAI-compatible APIs, with built-in inference engine for RAG, making it a powerful AI deployment solution.
  </Accordion>

  <Accordion title="What LLM providers does Open WebUI support?">
    Open WebUI supports:

    * Ollama (local models)
    * OpenAI-compatible APIs
    * LMStudio
    * GroqCloud
    * Mistral
    * OpenRouter
    * Anthropic
    * And many more OpenAI-compatible providers
  </Accordion>

  <Accordion title="Can Open WebUI run completely offline?">
    Yes! Open WebUI is designed to operate entirely offline. You can run local models using Ollama without any internet connection. Just set the `HF_HUB_OFFLINE=1` environment variable to prevent attempts to download models from the internet.
  </Accordion>

  <Accordion title="What are the system requirements?">
    **Minimum requirements:**

    * Python 3.11 or higher (for pip installation)
    * Docker (recommended for containerized deployment)
    * 2GB RAM minimum (more recommended for larger models)
    * Storage space for models and data

    **For GPU acceleration:**

    * Nvidia GPU with CUDA support
    * Nvidia CUDA container toolkit installed
  </Accordion>
</AccordionGroup>

## Installation & Setup

<AccordionGroup>
  <Accordion title="How do I install Open WebUI?">
    There are multiple installation methods:

    **Using pip:**

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

    **Using Docker:**

    ```bash theme={null}
    docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
    ```

    See the [Installation Guide](/getting-started/installation) for detailed instructions.
  </Accordion>

  <Accordion title="What ports does Open WebUI use?">
    * **Standalone/pip installation:** Port 8080 by default
    * **Docker installation:** Internal port 8080, typically mapped to external port 3000
    * **With --network=host:** Port 8080

    You can access Open WebUI at `http://localhost:3000` (Docker) or `http://localhost:8080` (pip/network=host).
  </Accordion>

  <Accordion title="How do I update Open WebUI?">
    **For Docker installations:**

    ```bash theme={null}
    docker stop open-webui
    docker rm open-webui
    docker pull ghcr.io/open-webui/open-webui:main
    # Run your docker run command again
    ```

    **For pip installations:**

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

    Check the [Updating Guide](https://docs.openwebui.com/getting-started/updating) for more details.
  </Accordion>

  <Accordion title="Should I use :main, :dev, or :ollama tags?">
    * **:main** - Stable release, recommended for production
    * **:dev** - Latest unstable features, use at your own risk
    * **:ollama** - Bundles Open WebUI with Ollama in one container
    * **:cuda** - Includes GPU acceleration support

    For most users, `:main` is recommended. Use `:dev` only if you want bleeding-edge features and can tolerate bugs.
  </Accordion>
</AccordionGroup>

## Features & Usage

<AccordionGroup>
  <Accordion title="What is RAG and how do I use it?">
    RAG (Retrieval Augmented Generation) allows you to provide documents to enhance your chat conversations. Open WebUI supports:

    * 9 vector database options
    * Multiple content extraction engines
    * Direct document upload to chat
    * Document library with `#` command access

    Simply upload documents to your chat or add them to your document library, then reference them using `#` before your query.
  </Accordion>

  <Accordion title="Can I use multiple models at once?">
    Yes! Open WebUI supports many-models conversations, allowing you to engage with various models simultaneously and leverage their unique strengths for optimal responses.
  </Accordion>

  <Accordion title="Does Open WebUI support voice and video calls?">
    Yes! Open WebUI includes hands-free voice and video call features using:

    * **Speech-to-Text:** Local Whisper, OpenAI, Deepgram, Azure
    * **Text-to-Speech:** Azure, ElevenLabs, OpenAI, Transformers, WebAPI

    This allows for dynamic and interactive chat environments.
  </Accordion>

  <Accordion title="How do I create custom models?">
    Use the Model Builder feature to:

    * Create Ollama models via the Web UI
    * Add custom characters/agents
    * Customize chat elements
    * Import models from [Open WebUI Community](https://openwebui.com/)
  </Accordion>

  <Accordion title="What authentication methods are supported?">
    Open WebUI supports:

    * Local authentication (email/password)
    * OAuth providers
    * LDAP/Active Directory
    * SSO via trusted headers
    * SCIM 2.0 for automated provisioning
  </Accordion>
</AccordionGroup>

## Administration

<AccordionGroup>
  <Accordion title="How do I manage user permissions?">
    Open WebUI includes granular Role-Based Access Control (RBAC):

    * Create detailed user roles and permissions
    * User groups for organized access management
    * Restrict model access and creation rights
    * Configure workspace sharing permissions
  </Accordion>

  <Accordion title="Can I use different databases?">
    Yes! Open WebUI supports:

    * SQLite (default, with optional encryption)
    * PostgreSQL
    * Cloud storage backends (S3, Google Cloud Storage, Azure Blob Storage)
  </Accordion>

  <Accordion title="How do I enable monitoring and observability?">
    Open WebUI has built-in OpenTelemetry support for:

    * Traces
    * Metrics
    * Logs

    This enables comprehensive monitoring with your existing observability stack.
  </Accordion>

  <Accordion title="Can Open WebUI scale horizontally?">
    Yes! Open WebUI supports:

    * Redis-backed session management
    * WebSocket support for multi-worker deployments
    * Multi-node deployments behind load balancers
  </Accordion>
</AccordionGroup>

## Troubleshooting

<AccordionGroup>
  <Accordion title="I'm getting connection errors to Ollama">
    This is often due to the Docker container not being able to reach the Ollama server. Use the `--network=host` flag:

    ```bash theme={null}
    docker run -d --network=host -v open-webui:/app/backend/data -e OLLAMA_BASE_URL=http://127.0.0.1:11434 --name open-webui --restart always ghcr.io/open-webui/open-webui:main
    ```

    See the [Troubleshooting Guide](/help/troubleshooting) for more solutions.
  </Accordion>

  <Accordion title="My data was lost after updating Docker">
    Always include the volume mount `-v open-webui:/app/backend/data` in your Docker command. This ensures your database is properly mounted and prevents data loss.
  </Accordion>

  <Accordion title="Where can I get support?">
    * [GitHub Issues](https://github.com/open-webui/open-webui/issues)
    * [Discord Community](https://discord.gg/5rJgQTnV4s)
    * [Open WebUI Documentation](https://docs.openwebui.com/)
  </Accordion>
</AccordionGroup>

## Enterprise & Licensing

<AccordionGroup>
  <Accordion title="Is there an enterprise version?">
    Yes! Open WebUI offers an Enterprise Plan with:

    * Custom theming and branding
    * Service Level Agreement (SLA) support
    * Long-Term Support (LTS) versions
    * And more!

    [Speak with our sales team](https://docs.openwebui.com/enterprise) for details.
  </Accordion>

  <Accordion title="What license does Open WebUI use?">
    Open WebUI uses a modified BSD 3-Clause License with an additional requirement to preserve "Open WebUI" branding. See the [LICENSE](https://github.com/open-webui/open-webui/blob/main/LICENSE) file for complete details.

    The branding restriction does not apply to:

    * Deployments with 50 or fewer users
    * Official contributors with written permission
    * Enterprise license holders
  </Accordion>
</AccordionGroup>