# Response speed

Response speed is a critical factor in the user experience when interacting with a conversational AI. AIsuru offers several options for optimizing response times across your Agents.

### Monitoring response times

Currently, AIsuru doesn't have a dedicated section specifically for monitoring response speed. However, you can observe your Agent's response times during user interactions or while running tests.

To monitor response times:

1. **Chat with your Agent or use the**  **Conversations section** to test and view response times;
2. **Take note of response times for different types of questions**.

### Optimizing response speed

If your Agent's response times are slower than you'd like, there are several strategies you can adopt to improve performance. Let's walk through them 👇

#### 1. Switch the language model

Some language models are faster than others. To change the model:

* Go to your Agent's "Settings" section;
* Select the "Generative AI" tab;
* [**Choose a faster model**](/en/generative-ai/providers.md) from the available options.

{% hint style="warning" %}
Switching the language model can improve response speed, but it may also change how the Agent interacts with users or lower the quality of responses. After switching models, **make sure you test your Agent thoroughly**!
{% endhint %}

#### 2. Trim the instructions

Long, complex instructions can slow down response times. To streamline them:

* Go to your Agent's "Settings" section;
* Access the AI section;
* Edit the AI "Instructions", keeping only the essential ones;
* If appropriate, consider emphasizing that the Agent should send concise messages to users;
* Save the changes.

#### 3. Optimize the knowledge base

An overly large or poorly organized knowledge base can slow down responses. Consider:

* Reviewing and consolidating content in the "Contents" section;
* Removing duplicate or unnecessary information;
* Better organizing content using [contexts](/en/agent-training/advanced/content-customization/context-topics-and-labels.md), [personas](/en/agent-training/advanced/content-customization/customizing-responses-by-date-and-location.md), and [question variants](/en/agent-training/contents/manual-content-creation.md).

#### 4. Use Expert Groups

If your Agent handles a wide range of topics, consider using the [Expert Group](/en/extensions/expert-group.md) feature:

* Create several specialized Agents for specific areas;
* Set up an Expert Group to route questions to the most appropriate Agents;
* This approach can significantly improve response times for specialized questions.

Remember that **optimizing response speed is an iterative process**. After making changes, keep monitoring performance and adjusting as needed. Find the right balance between speed and response quality to deliver the best possible experience to your users.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.aisuru.com/en/analytics/agent-analysis/response-speed.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.