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

# GenAI Features - Contact Center AI

<Badge icon="arrow-left" color="gray">[Back to Generative AI Features](/ai-for-service/generative-ai-tools/genai-features)</Badge>

LLM-powered features for Contact Center AI that improve agent response quality, conversation summaries, disposition accuracy, and real-time sentiment visibility.

## Model Feature Matrix

(✅ Supported | ❌ Not supported)

| Feature                                  | Azure OpenAI        | OpenAI              | Custom LLM | XO GPT |
| ---------------------------------------- | ------------------- | ------------------- | :--------: | :----: |
| Agent Response Rephrasing                | ❌                   | ❌                   |      ❌     |    ✅   |
| Conversation Summary                     | ❌                   | GPT-3.5             |      ✅     |    ❌   |
| Disposition Prediction for Agent Wrap-Up | GPT-4o, GPT-4o mini | GPT-4o, GPT-4o mini |      ✅     |    ❌   |
| Sentiment Analysis                       | GPT-4o              | GPT-4o              |      ✅     |    ❌   |

***

## Features

### Agent Response Rephrasing

Rephrases agent responses in the following tones:

| Tone          | Description                                                  |
| ------------- | ------------------------------------------------------------ |
| **Formalize** | Makes the response suitable for business or formal contexts. |
| **Friendly**  | Promotes an approachable and friendly tone.                  |
| **Expand**    | Expands the response for formal settings.                    |
| **Rephrase**  | Elevates formality for business or professional contexts.    |

### Conversation Summary

Generates LLM-based conversation summaries at agent handoff, transfer between agents, and conversation wrap-up. [Learn more](/ai-for-service/contact-center/agent-console#arrival-summary).

### Disposition Prediction for Agent Wrap-Up

Analyzes the full conversation transcript and disposition metadata, then recommends the most appropriate wrap-up code. Agents can accept or override the recommendation, improving accuracy and reducing wrap-up time. [Learn more](/ai-for-service/contact-center/agent-console#dispositions).

### Sentiment Analysis

Displays a real-time sentiment indicator during active chat conversations, updating dynamically as the conversation progresses. [Learn more](/ai-for-service/contact-center/agent-console#real-time-sentiment).

***