---
title: 2. Data Training Setup
slug: 2-data-training-setup
docTags: 
createdAt: 2026-03-23T16:58:11.072Z
---

After configuring the **Gen AI model**, the next step is to train the agent with the data it will use to answer user queries.

The agent can be trained using different types of data sources such as **documents, website URLs, or structured scripts**. This training allows the agent to retrieve relevant information when responding to users.


**The following methods can be used to train the agent:**

### 3.1  [URL Training](docId\:llgHtI4RPTcZ6AzFuaWXl)

- Train the agent using content from website URLs. 
- The agent will extract and learn information from the provided web pages. 

### 3.2 [RAG Document Training](docId\:RHBC-7jYc9umK9LmnLplt)

Upload documents such as **PDFs or files** to train the agent.

- The agent uses these documents to retrieve information and generate responses during conversations. 

### 3.3 [Sitemap Training](docId\:QdBWIbMGWGMN_7TbvH8kZ)

Train the agent using **multiple pages from a website sitemap**:

- This allows the agent to learn content from an entire website structure. 

### 3.4 [Custom Script Training](docId\:kXGOANFOcA2uUWx9iUBze)

- Train the agent using **structured scripts** for specific use cases. 
- This method is useful when responses need to follow a predefined format or workflow.

##  📚 Next Steps

Once the data training is completed, continue with [3. Path Setup](docId\:G-jippCSr0vJWz4rIchMf) to define how the agent navigates conversations and handles different user scenarios.