> For the complete documentation index, see [llms.txt](https://docs.sprinthub.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.sprinthub.com/en/topics/artificial-intelligence/ai-agents/legacy-ai-agents-notice/chatgpt-models/chatgpt-models-website-and-text.md).

# ChatGPT Models - Website and Text

<figure><img src="/files/b29339c88c7fca7624f61ccde49ec92c784bcfdb" alt=""><figcaption></figcaption></figure>

## In the website GPT chat model, you can extract information from a website of your choice and incorporate it into ChatGPT in SprintHub. This allows you to associate it with a chatbot and provide specific information about that site. In addition, in text mode, you can use a text of your choice to extract information and pass it on to a ChatBot as well.

1. Define a name for your Model.
2. Create a description for your model.
3. A higher level of contextualization means that Chat GPT will receive more information from the company during a conversation, increasing the quality of the responses. On the other hand, response time may increase, as well as the tokens used. For a low level of contextualization, an estimated cost of around 300 to 400 tokens per message is expected. For a medium level of contextualization, an estimated cost of around 600 to 700 tokens per message is expected. For a medium level of contextualization, an estimated cost of around 1000 to 1200 tokens per message is expected.
4. Select which Chat GPT model you want to use.
5. Increasing the temperature, such as to 1, will make Chat GPT more creative and prone to random responses. Decreasing the temperature, such as to 0, will make the model more cautious and conservative in its responses.
6. Add your new questions and answers. Enter the questions you want to "simulate", as if a customer were asking them, and the answers you want Chat GPT to provide, aligned with your company's guidelines.
7. It is possible to import spreadsheets in CSV and XLSX formats that contain predefined question-and-answer templates that you have already configured.
8. In additional information, you can describe the personality and role of the virtual "agent" in your Chat GPT model. For example, you can define it as a friendly company attendant, whose role is to respond exclusively with information that has been previously defined by you.

### To configure your Chat GPT in the mode **Text** and in the mode **Website** follow the step-by-step guide:&#x20;

Click on **Convert** and then on '**Models** **Chat GPT**'

<figure><img src="/files/7445ad917c99c62bb1de7c12f2118de90fb3dfb9" alt=""><figcaption></figcaption></figure>

When you start creating your Chat GPT model, click on '**Website**';

<figure><img src="/files/859700fd2c21a3b573480a301fb619fdd90fb422" alt=""><figcaption></figcaption></figure>

Enter the **URL** of the website you want to **extract the information**;

<figure><img src="/files/dc67690f5e38414f4cf26746bdce7cb4fade747f" alt=""><figcaption></figcaption></figure>

Then click on '**Extract content**';

{% hint style="warning" %}
**ChatGPT charges a fee per extracted file.**
{% endhint %}

{% embed url="<https://openai.com/api/pricing/>" %}

<figure><img src="/files/18c981351be4c2ceec068f981ae79dcc92fd0628" alt=""><figcaption></figcaption></figure>

After clicking Extract content, wait a few minutes, and shortly after the website content will be ready.

<figure><img src="/files/966f0df33184f6c7836c38f60db7e7cce11d3134" alt=""><figcaption></figcaption></figure>

Click on **Save and Close** after **configuring** your model.

To **test** your Chat GPT model, go back to the general Chat GPT models tab, and click on the name of the model you want to test;

<figure><img src="/files/7dd159a6c7118ab9985ced5f6861bda019c1b914" alt=""><figcaption></figcaption></figure>

Ask any questions you want about the topics **related to the content extracted from the website**;

<figure><img src="/files/7b635f1d458ae84a449c35979b7af60026735bcd" alt=""><figcaption></figcaption></figure>

Note that next to the test chat, the recent activities of the Chat GPT model appear. And in it, there is information about the amount that the **CHATGPT** **charges** for each trained (extracted) model.

<figure><img src="/files/9499cefe4e346487919f992e2147c8bda6e2d13c" alt=""><figcaption></figcaption></figure>

***

To configure the text mode, click on **'Text'**;

<figure><img src="/files/0a598b8c1e4ac2d1ec32b05900903834315534b8" alt=""><figcaption></figcaption></figure>

Copy or type the text of your preference.   &#x20;

<figure><img src="/files/8295f891c90cb4526840fa480893309887a2ad0e" alt=""><figcaption></figcaption></figure>

Click on **Save and Close** after **configuring** your model.

Just as in the Website form mentioned above, you can **test** your text model;&#x20;

<figure><img src="/files/de5e55b258cbeee8583b45c03259c8877ba04a0b" alt=""><figcaption></figcaption></figure>

*After testing the Chat GPT models, they will be ready for use on the platform and can be used in ChatBots, for example.*

<mark style="color:purple;">**See articles on similar topics:**</mark>

{% content-ref url="/pages/b22f52d8c3fd81e06678b8e50278dbe2790712b3" %}
[Chatbot - Using AI Agents](/en/topics/convert/chatbot/chatbot-common-questions/chatbot-using-ai-agents.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://docs.sprinthub.com/en/topics/artificial-intelligence/ai-agents/legacy-ai-agents-notice/chatgpt-models/chatgpt-models-website-and-text.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.