> ## Documentation Index
> Fetch the complete documentation index at: https://knowledge.bitbybit.studio/llms.txt
> Use this file to discover all available pages before exploring further.

# Set up your AI agent in AI Studio

> Create an AI agent in AI Studio and complete the current Identity setup, including brand information, AI model, personality, and instructions.

Use this page to set up the **Identity** tab in **AI Studio**.

If you need an overview of the full workspace first, start with [Get started with AI Studio](/ai-studio/getting-started-with-your-ai-agent).

## Prerequisites

* You can access **AI Studio**.
* You have the brand information you want your AI agent to use.

## Step 1: Open AI Studio

* Go to **AI Studio**.
* Open **AI Studio**.
* Create a new agent if you are starting from scratch.

## Step 2: Complete Brand information

In **Brand information**, enter the core business details the AI agent should use.

* Fill in **Brand name**.
* Fill in **Brand description**.

## Step 3: Choose an AI model

In **AI model**, pick the agent best suited for your customers' needs. You can change it at any time.

The picker shows two options:

* **Conversation** — optimized for fast, accurate replies. Best for FAQs, order tracking, and product questions.
* **Reasoning** — uses deeper reasoning for complex, multi-step questions. Best for troubleshooting and special requests. Available on the **Standard** plan and above; on lower plans the option is locked with an upgrade prompt.

<Tip>
  **Custom model** (Pro Bundle plan and above, with the Custom model feature enabled): a separate **Model** dropdown appears so you can pin the agent to a specific underlying LLM. The dropdown currently exposes:

  * Gemini family — **Gemini 3 Flash**, **Gemini 3 Pro**, **Gemini 3.1 Flash-Lite**, **Gemini 3.1 Pro**, **Gemini 2.5 Pro**, **Gemini 2.5 Flash**.
  * OpenAI family — **GPT-5.1**, **GPT-5**, **GPT-5 Chat**, **GPT-5 Mini**, **GPT-5 Nano**, **GPT-4o**, **GPT-4.1**, **GPT-4.1 Mini**, **GPT-4.1 Nano**.

  When you select a **GPT-5** or **Gemini 3** model, a **Reasoning effort** dropdown appears so you can tune how much the model deliberates before replying. Higher effort improves quality on complex questions but increases latency and cost.
</Tip>

## Step 4: Complete Personality

In **Personality**, choose how the AI agent communicates with customers.

* Choose **Tone of voice**.
* Choose **Response style**.

<Note>
  The AI agent automatically replies in the customer's language, so there is no language setting on this tab. To make the agent always reply in a specific language, add a rule in the **Instructions** field (Step 5) — for example, "Always respond in English, regardless of the language the customer uses."
</Note>

## Step 5: Add Instructions

In **Instructions**, describe what your AI agent should do. Use this field only for rules that should consistently shape the agent's behavior.

* Enter the instructions you want the AI agent to follow.
* Optionally click **Improve instruction** to rewrite your draft with AI assistance.

<Tip>
  The **Instructions** field accepts up to **15,000 characters**. If you exceed the cap, the UI blocks save — split long rule sets across the Brand information, Personality, and Skills surfaces instead of stuffing everything into Instructions.
</Tip>

## Step 6: Save your AI agent

* Review the fields in **Brand information**, **AI model**, **Personality**, and **Instructions**.
* Save your changes.

## Related articles

* [Get started with AI Studio](/ai-studio/getting-started-with-your-ai-agent)
* [Add and manage skills in AI Studio](/ai-studio/how-to-setup-skillset-in-ai-studio)
* [Best practice to prompt your AI agent](/ai-studio/best-practice-to-prompt-your-ai-agent)