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

# AI Models

> Deep dive on connecting different model providers

GAIA supports multiple AI providers through a unified interface.
You will need 2 main AI Models in this setup

* AI Provider / LLM (Language model ): which will be used in chatting process
* Emebdding Provider:  which will be used in Indexing process

## AI Provider Setup

<Tabs>
  <Tab title="Ollama (Local)">
    <Steps>
      <Step title="Install Ollama">
        Download from [ollama.com](https://ollama.com/download) and start the service:

        ```bash theme={null}
        ollama serve
        ```
      </Step>

      <Step title="Configure in GAIA AI">
        <img className="dark:hidden mx-auto" src="https://mintcdn.com/tuhh-d1de0e06/NLF4OVAebw6tUaJP/images/ollama-view-light.png?fit=max&auto=format&n=NLF4OVAebw6tUaJP&q=85&s=6b6b9fa3d16f4e15fb5be67ef2626557" alt="Ollama setup" style={{ width:"99%", borderRadius: "12px",  }} width="1282" height="797" data-path="images/ollama-view-light.png" />

        <img className="hidden dark:block" src="https://mintcdn.com/tuhh-d1de0e06/NLF4OVAebw6tUaJP/images/ollama-view-dark.png?fit=max&auto=format&n=NLF4OVAebw6tUaJP&q=85&s=38fb9d4baf8f31cc0eef1820cff81723" alt="Ollama setup" style={{ width:"99%", borderRadius: "12px",  }} width="1282" height="797" data-path="images/ollama-view-dark.png" />

        * Go to **Credentials → AI Providers**
        * Select **Local Tab**
        * Select **Ollama** provider
        * Set URL:
          * Local: `http://localhost:11434`
          * Docker: `http://host.docker.internal:11434`
        * pull your model (e.g., `llama3`)
        * Done
      </Step>
    </Steps>
  </Tab>

  <Tab title="Cloud Providers">
    **No installation needed!**

    Simply get your API key from your preferred provider:

    * OpenAI
    * Anthropic
    * Google AI
    * And more...

    Add it in **Settings → AI Models**
  </Tab>

  <Tab title="OpenAI-Compatible">
    **For LM Studio, vLLM, LocalAI, etc.**

    <div className="overflow-hidden">
      <img className="dark:hidden mx-auto" src="https://mintcdn.com/tuhh-d1de0e06/NLF4OVAebw6tUaJP/images/openai-view-light.png?fit=max&auto=format&n=NLF4OVAebw6tUaJP&q=85&s=49e10bc0d27156e9b55a05745dd4bea0" alt="OpenAI-compatible setup" style={{ width:"100%", borderRadius: "12px", marginLeft:"10px" }} width="1282" height="797" data-path="images/openai-view-light.png" />
    </div>

    <img className="hidden dark:block" src="https://mintcdn.com/tuhh-d1de0e06/NLF4OVAebw6tUaJP/images/openai-view-dark.png?fit=max&auto=format&n=NLF4OVAebw6tUaJP&q=85&s=9c15536d23dc225130f58113408b6744" alt="OpenAI-compatible setup" style={{ width:"99%", borderRadius: "12px",  }} width="1282" height="797" data-path="images/openai-view-dark.png" />

    You'll need:

    * API URL
    * API Key (if required)

    Configure in **Settings → AI Models**
  </Tab>
</Tabs>

Same process will be needed for embedding provider