> For the complete documentation index, see [llms.txt](https://docs.aisuru.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aisuru.com/en/generative-ai/activate-generative-ai-and-change-model.md).

# Activate generative AI and change the model

Generative AI makes your Agents smarter and more dynamic: once enabled, your Agent will adapt its responses based on the conversation and independently understand the context of each request.

### Generative AI: when to use it

Generative AI is particularly useful for:

* **Generating more natural responses** that fit the context;
* **Dynamically adapting** to different user requests;
* **Providing more detailed and relevant information**;
* **Personalizing the tone and style** of responses based on context.

{% hint style="warning" %}
Generative AI **can also give users incorrect information**. AIsuru offers many tools to minimize generative AI errors, but it's always possible that a response may be wrong or partially wrong.

Response quality depends heavily on the [provider and model you choose](/en/generative-ai/providers.md) and the architecture of your Agent.
{% endhint %}

### How to enable generative AI for your Agent

Generative AI is automatically enabled for every new Agent. In case it is not, you can enable it following these simple steps:

1. Open the management panel for the Agent you want to configure;
2. In the sidebar, click **Settings**;
3. Inside the Settings page, click on the **AI** section;
4. Under "Response mechanism" select:
   1. Contents first, then AI: if you want to use GenAI only when there are no useful [content](/en/agent-training/contents.md) that could be used to directly address the user's questions.
   2. AI only: if you want to always use GenAI to answer user's questions. Note that contents will still be used, but only to provide more context to the AI model chosen.

✨ Congratulations! You've just enabled generative AI for your Agent.

{% hint style="info" %}
You can also select Contents only as response mechanism. In this case, GenAI will be completely disabled and your Agent will answer user's questions only when AIsuru finds a good match between the question and the contents of your Agent.
{% endhint %}

### How to change your Agent's model

1. **Q\&A configuration**: this is the main model that:
   1. Reads and follows instructions;
   2. Analyzes the most relevant [contents](/en/agent-training/contents.md) in your Agent;
   3. Generates responses to user questions;
   4. Decides when to use functions or dynamic intents.
2. **Import/export configuration**: lets you change the model that handles [importing your text documents](/en/agent-training/document-import/importing-text-documents.md);
3. [**Deep Thinking**](/en/extensions/deep-thinking.md) **configuration**: lets you change the model that decides what to remember from conversations with each user (only active when Deep Thinking is enabled).

Below Response mechanism, you can find a dropdown menu **Choose model**. In this menu, you can choose your preferred model to:

* Read and follow the instructions of the Agent;
* Read the most relevant contents of your Agent;
* Generate responses to the user's questions;
* Decide to use MCP, skills, and functions.

AIsuru automatically selects the default configuration for these tasks, which generally corresponds to the latest Claude Sonnet model.

After enabling generative AI and configuring your models, you can:

* Edit the [**instructions**](/en/generative-ai/ai-instructions.md) of your Agent;
* Enable or disable [**Deep Thought**](/en/extensions/deep-thinking.md);
* Configure [**other generative AI settings**](/en/generative-ai/advanced/other-generative-ai-settings.md).