> For the complete documentation index, see [llms.txt](https://docs.dapta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapta.ai/ai-voice-agents/how-to-set-up-your-ai-voice-agent/dapta-auto.md).

# Dapta Auto

Dapta Auto is the model Dapta chooses and keeps up to date for your Voice Agent. New Voice Agents start on it, and you can pick it under LLM Model for the ones you have; the per-minute price of a call

**Dapta Auto** is the AI model Dapta manages for you. The language model is what your Voice Agent uses to think and answer during the call. Instead of comparing the list and checking every few weeks whether a better one came out, choose Dapta Auto once: Dapta picks the model behind it for fast, efficient answers and moves it to a better one after testing it. Your agent and its prompt stay as they are.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCy5rSNtQmtqYCGzJlNEB%2Fuploads%2Fgit-blob-9b0e36cb88d033c4f5bfb205fc371a7fe6ecda2e%2Fvoice-agents-dapta-auto.mp4?alt=media>" %}

{% file src="/files/sH14XFAm3aiFLQmb645h" %}

## What changes for you

|                       |                                                                                                                                     |
| --------------------- | ----------------------------------------------------------------------------------------------------------------------------------- |
| **New Voice Agents**  | Start on Dapta Auto, whether you create them yourself or with Dapti. This is the case since 20 October 2026.                        |
| **Your Voice Agents** | Keep the model they use today. Moving one to Dapta Auto is a choice you make per agent, with the steps below.                       |
| **Price**             | The same per-minute rate as any other model. See [How credits work](/my-account/how-credits-work-in-dapta.md).                      |
| **Reasoning effort**  | Not shown for Dapta Auto. Dapta sets it, so the agent answers without the pauses that high reasoning adds to a call.                |
| **Updates**           | When Dapta moves Dapta Auto to a better model, every agent on it gets the change. You don't touch anything.                         |
| **Availability**      | Every Voice Agent can use it. If it isn't in your list, see [I don't see Dapta Auto](#i-dont-see-dapta-auto-in-the-llm-model-list). |

***

**📘 What you'll learn in this document:**

* **Step 1:** Find the **LLM Model** setting of your agent.
* **Step 2:** Choose **Dapta Auto**.
* **Step 3:** Save and test it.

***

#### Step 1: Open the LLM Model setting

1. Open your **Voice Agent** and go to the **Settings** tab.
2. Open **Voice Settings** and find **LLM Model**, the AI model your agent uses to think and respond during calls.

***

#### Step 2: Choose Dapta Auto

Click the list. **Dapta Auto** is the first option.

<figure><img src="https://3835013762-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCy5rSNtQmtqYCGzJlNEB%2Fuploads%2Fgit-blob-20453ebcae0c580dc938e0086e4f223359aa5c65%2Fvoice-agents-dapta-auto-01-model-list.png?alt=media" alt="The LLM Model list of a Voice Agent with Dapta Auto first"><figcaption></figcaption></figure>

Once it's selected, the field reads *Managed by Dapta for fast, efficient responses. Updated automatically.*, and the **Reasoning effort** setting goes away: Dapta handles it.

<figure><img src="https://3835013762-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCy5rSNtQmtqYCGzJlNEB%2Fuploads%2Fgit-blob-0ea66dc82bcd7a04eacc6b57ca7edd379504dd31%2Fvoice-agents-dapta-auto-02-selected.png?alt=media" alt="Dapta Auto selected in the LLM Model field, with its description below"><figcaption></figcaption></figure>

***

#### Step 3: Save and test it

1. Click **Save changes**.
2. Open **Test Agent** and try it with a **Test Chat** or a **Web Call**, using the questions your callers ask most.

{% hint style="info" %}
The prompt still decides how your agent behaves. A new model can phrase things differently, so listen to a few test calls before the agent takes real ones, and adjust the prompt if you need to.
{% endhint %}

***

## Which model is behind Dapta Auto?

Dapta doesn't show it, on purpose. It is the part Dapta takes care of, and it changes when Dapta finds a better option for calls. What doesn't change is what you see: the name **Dapta Auto** and the price of your calls.

If you need a specific model by name, for example because your company approves only certain AI providers, choose that model from the list instead.

## Troubleshooting

<details>

<summary>I don't see Dapta Auto in the LLM Model list</summary>

Dapta Auto runs on Dapta's voice infrastructure, where Voice Agents run today. If your workspace is still on the previous one, the option doesn't appear. Contact support and we'll tell you where you stand.

</details>

<details>

<summary>My agent sounds different after the switch</summary>

Make a few test calls and adjust the prompt, or use **Improve with AI**. If you'd rather go back, choose the previous model in **LLM Model** and save: nothing else on the agent changed.

</details>

## Related pages

{% content-ref url="/pages/dAhoJK7K0MZ80um6VhFo" %}
[Test Agent](/ai-voice-agents/creating-a-voice-agent/test-agent.md)
{% endcontent-ref %}

{% content-ref url="/pages/E4aI8Qcgr3XygWSSNesf" %}
[Improving prompts with AI](/ai-voice-agents/how-to-set-up-your-ai-voice-agent/improving-prompts-with-ai.md)
{% endcontent-ref %}

{% content-ref url="/pages/U29bCu8KL35PfZDKEqD1" %}
[Dapta Auto](/ai-text-agents/setting-up-your-text-agent/dapta-auto.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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.dapta.ai/ai-voice-agents/how-to-set-up-your-ai-voice-agent/dapta-auto.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.