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

# Model

The **Model** tab allows you to choose the Large Language Model (LLM) that powers your AI agent. The selected model directly affects response quality, speed, context size, reasoning capability, and credit consumption.

<Frame>
  <img src="https://mintcdn.com/chatzyai/ROvnVnZiw9dZ31Hu/ai-agents/conversational-ai-agent/settings/model.png?fit=max&auto=format&n=ROvnVnZiw9dZ31Hu&q=85&s=9343a083ceaa2c42af3a662a6ac41b91" width="1639" height="550" data-path="ai-agents/conversational-ai-agent/settings/model.png" />
</Frame>

***

## Temperature

The **Temperature** setting controls how deterministic or creative the AI's responses are.

* **Reserved (0)** – Produces factual, consistent, and predictable responses. Recommended for customer support, FAQs, lead qualification, and business workflows.
* **Creative (>0)** – Generates more diverse and creative responses. Ideal for brainstorming, content generation, and open-ended conversations.

***

## Available Models

ChatzyAI supports models from multiple AI providers.

### OpenAI

* **GPT-4.1 1M** – High-quality model with a 1 million token context window.
* **GPT-4.1 Mini 1M** – Faster and more cost-efficient version of GPT-4.1.
* **GPT-4o 128K** – Fast, multimodal model suitable for most conversational tasks.
* **GPT-4o Mini 128K** – Lightweight version of GPT-4o for high-volume workloads.
* **GPT-5 400K** – Advanced reasoning model with a large context window.
* **GPT-5 Mini 400K** – Faster and lower-cost version of GPT-5.
* **GPT-5.4 1M** – Large-context model designed for complex workflows.
* **GPT-5.4 Mini 272K** – Cost-efficient GPT-5.4 variant.
* **GPT-5.6 Luna 1M** – Optimized for high-quality conversational experiences.
* **GPT-5.6 Terra 1M** – Enterprise-grade model with excellent reasoning and context handling.
* **O4 Mini 200K** – Optimized reasoning model for analytical and multi-step tasks.

### Anthropic

* **Claude Haiku 4.5 200K** – Extremely fast and efficient for lightweight conversations.
* **Claude Sonnet 4.6 1M** – Powerful reasoning model with a large context window.

### Google

* **Gemini 2.5 Flash 1M** – Balanced model offering excellent speed and reasoning.
* **Gemini 2.5 Flash Lite 1M** – Lightweight and cost-efficient Flash model.
* **Gemini 3.1 Flash Lite 1M** – Faster generation for high-volume use cases.
* **Gemini 3.5 Flash 1M** – Google's latest Flash model with improved reasoning and performance.

***

## Choosing the Right Model

Different models are optimized for different workloads.

| Use Case                   | Recommended Models                                             |
| -------------------------- | -------------------------------------------------------------- |
| Customer Support           | GPT-4o, GPT-4o Mini, Claude Haiku 4.5                          |
| Lead Qualification         | GPT-4.1 Mini, GPT-4o Mini, Gemini Flash Lite                   |
| Sales Conversations        | GPT-5, GPT-5.4, Claude Sonnet 4.6                              |
| Complex Reasoning          | GPT-5, GPT-5.6 Terra, Claude Sonnet 4.6, O4 Mini               |
| Large Knowledge Bases      | GPT-4.1, GPT-5.4, GPT-5.6 Luna, Gemini Flash                   |
| Cost-Optimized Deployments | GPT-4o Mini, GPT-4.1 Mini, Claude Haiku 4.5, Gemini Flash Lite |

***

> **Tip:** Larger context windows allow the AI to reference more conversation history and larger knowledge bases, but may consume more credits depending on the selected model.