> ## Documentation Index
> Fetch the complete documentation index at: https://docs.20x.business/llms.txt
> Use this file to discover all available pages before exploring further.

# LLM Tab

> LLM provider and model selection for conversational intelligence

The **LLM Tab** in Agent Studio configures the underlying Large Language Model (LLM) that powers your agent's comprehension, reasoning, tool execution, and response generation.

***

## Settings Overview

| Field        | Description                                                   | Requirement | Options                                  |
| ------------ | ------------------------------------------------------------- | ----------- | ---------------------------------------- |
| **Provider** | The LLM provider hosting the language model                   | Required    | `OpenAI`, `Google`                       |
| **Model**    | The specific model used for reasoning and generating dialogue | Required    | Available models for the chosen provider |

***

## Detailed Field Specifications

### Provider

* **Description:** Selects the AI platform provider for your agent's language intelligence.
* **Supported Options:**
  * **OpenAI** — Industry-standard models with strong reasoning, instruction following, and tool-calling fidelity.
  * **Google** — High-speed, multimodal Gemini models optimized for ultra-low latency and complex contextual understanding.
* **Prerequisites:** Provider credentials must be connected under [Integrations](/features/integrations) or enabled for your tenant. Selecting a provider dynamically populates the available models in the **Model** selector.

### Model

* **Description:** Specifies the exact language model version that processes transcripts and generates the agent's real-time replies.
* **Supported Models:**
  * **OpenAI:** `GPT-4o`, `GPT-4o mini`, `GPT-4 Turbo`
  * **Google:** `Gemini 1.5 Pro`, `Gemini 1.5 Flash`, `Gemini 2.0 Flash`
* **Selection Guidance:**
  * **Low Latency / High Volume:** Models like `GPT-4o mini` or `Gemini 1.5 Flash` provide sub-second responses ideal for fast, conversational voice interactions.
  * **Complex Reasoning / Deep Tool Use:** Models like `GPT-4o` or `Gemini 1.5 Pro` are recommended for agents handling multi-step workflows, complex tool schemas, or strict domain logic.

***

## How to Configure

<Steps>
  <Step title="Open Agent Studio">
    Navigate to **Agents** in the dashboard and select the agent you want to configure.
  </Step>

  <Step title="Select LLM Tab">
    Click on the **LLM** tab in the configuration panel.
  </Step>

  <Step title="Choose Provider & Model">
    Select your desired **Provider** from the dropdown, then select the target **Model**.
  </Step>

  <Step title="Save Changes">
    Click **Save Changes** to immediately apply the new model to your agent.
  </Step>
</Steps>