> For the complete documentation index, see [llms.txt](https://partners.botpenguin.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://partners.botpenguin.com/partner-platform-guidelines/configurations/default-ai-model-configuration.md).

# Default AI Model Configuration

The **Default AI Model** section allows workspace administrators to configure AI and Voice capabilities for their customers — including model selection, usage limits, credit purchases, key integrations, and advanced features like image scraping and voice usage.

***

### Overview

As a partner/workspace administrator, this section gives you control over:

* Which AI model powers your customers' interactions
* How many AI message credits customers receive by default
* Whether customers can purchase additional credits or bring their own API keys
* Voice usage settings and minute allocations
* Advanced capabilities like image scraping

***

### AI Settings

#### Select AI Model

Choose the default AI model that will be used for all customer interactions in your workspace. The model you select here determines which AI engine powers the **free messages** included with customer plans.

| Option        | Underlying Model       |
| ------------- | ---------------------- |
| ChatGPT / GPT | GPT-4o mini (standard) |
| Gemini        | Gemini standard model  |
| Claude        | Claude standard model  |

{% hint style="info" %}
The model you configure here is specifically tied to the free AI messages allocated to customers on any plan. This is not a general-purpose selection — it defines the engine behind those complimentary credits.
{% endhint %}

***

#### Default Account Limit

Set the default number of AI message credits allocated to each customer account (e.g., `100`).

This limit helps you control operational costs and prevent excessive usage. Customers who hit this limit will either need to purchase additional credits (if enabled) or wait for a reset.

{% hint style="info" %}
All changes are applied post the "Update" button being clicked
{% endhint %}

***

#### AI Configuration Options

**Allow customer to purchase AI message credits**

When enabled, customers can self-serve additional AI usage by purchasing message credit add-ons once their default allocation runs out. When disabled, usage is strictly capped at the allocated amount.

<figure><img src="/files/5epGdHVAsThbOYceYOrA" alt=""><figcaption></figcaption></figure>

***

**Allow customer to integrate their own API keys**

When enabled, customers can connect their own AI provider API keys for billing and usage directly with the provider.

{% hint style="warning" %}
**Important limitation:** This option is not available in isolation. Customers cannot *only* use their own key — they must either purchase an add-on or have an existing add-on active in order to integrate their own key. If a customer only wants to bring their own key without purchasing an add-on, that is **not currently supported**.
{% endhint %}

***

**Allow Image Scraping**

Enables the AI to analyze and extract information from images. Enable this only if your customers require it, as it may increase resource usage and may require a higher plan tier.

***

### Voice Settings

#### Allow customer to purchase Voice minute credits

Enables customers to purchase additional voice usage minutes beyond their default allocation. This is ideal for use cases involving voice bots, IVR systems, or call-based AI interactions.

Voice credits are specifically used for **voice minutes** and **voice allocation** — they are separate from AI message credits and managed independently.

<figure><img src="/files/uQwnS6LW1mOqU2PYoONm" alt=""><figcaption></figcaption></figure>

***

#### Allow customer to integrate their own Voice provider keys

When enabled, customers can connect their own voice service provider keys, giving them direct provider billing and greater customization. Similar to AI key integration, this works in conjunction with an add-on purchase.

***

### Troubleshooting

<details>

<summary><strong>AI messages are not working</strong></summary>

The most common cause is that the configured API key has exhausted its credentials or quota.

Check whether:

* The API key is still active.
* The provider account has sufficient balance/quota remaining.
* The selected AI provider is operational.

</details>

<details>

<summary><strong>Voice minutes not being allocated correctly</strong></summary>

Voice settings are managed separately from AI message settings.

Verify that:

* Voice credits have been purchased or allocated.
* Voice configuration settings have been saved.
* The **Update** button was clicked after making changes.

</details>

<details>

<summary><strong>Customer cannot integrate their own API key</strong></summary>

This is a known platform limitation.

Customers cannot use the **Integrate Own Key** option unless they have also purchased an add-on.

Currently there is no standalone **Bring Your Own Key (BYOK)** option.

Customers must either:

1. Purchase an add-on, which unlocks API key integration.
2. Use the platform's default AI credits.

</details>

***

### Frequently Asked Questions

<details>

<summary><strong>What is the purpose of the AI model integration in this section?</strong></summary>

The integration in this section is specifically used to power the free AI messages included with customer plans.

The selected model determines which AI engine serves those complimentary messages.

This setting is **not** a global AI model configuration for all platform features.

</details>

<details>

<summary><strong>What model is actually used when I select ChatGPT, Gemini, or Claude?</strong></summary>

Current provider mappings:

* **ChatGPT / GPT** → GPT-4o mini
* **Gemini** → Standard Gemini model
* **Claude** → Standard Claude model

These mappings may change over time. Contact support for the latest model assignments.

</details>

<details>

<summary><strong>Can a customer use only their own API key without purchasing anything?</strong></summary>

No.

The **Integrate Own Key** feature is only available when an add-on has been purchased.

Customers must either:

* Purchase an add-on to unlock API key integration, or
* Use the platform's included AI credits.

Standalone API key integration is not currently supported.

</details>

<details>

<summary><strong>What are Voice credits used for?</strong></summary>

Voice credits are used exclusively for voice minutes.

These credits cover the time during which a customer's voice bot or IVR system is actively being used.

Voice credits are separate from AI message credits and must be purchased and managed independently.

</details>

<details>

<summary><strong>Can I set different limits for different customers?</strong></summary>

The **Default Account Limit** applies globally to all new customers within your workspace.

For customer-specific limits:

* Review the individual customer's account settings.
* Contact support if additional customization is required.

</details>

***

### Best Practices

* Set conservative default credit limits to prevent unexpected usage spikes.
* Enable **credit purchases** to reduce support load — customers can self-serve when they run out.
* Only enable **custom key integration** for trusted or enterprise-tier customers, as it requires an active add-on.
* Enable **image scraping** only when your customer use cases explicitly require it, since it adds resource overhead.
* Configure **voice settings** only if your customers are actively using voice bots or call-based features.

***

### Common Scenarios

| Customer Type             | Recommended Setup                                           |
| ------------------------- | ----------------------------------------------------------- |
| Small / starter customers | Low default limit, credit purchases disabled                |
| Growing teams             | Credit purchases enabled, default limit moderate            |
| Enterprise users          | Custom key integration enabled (with add-on), higher limits |
| Voice bot customers       | Voice credits enabled + voice provider key integration      |

***

### Need Help?

For questions or concerns, reach out to the BotPenguin support team at **<support@botpenguin.com>**.


---

# 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:

```
GET https://partners.botpenguin.com/partner-platform-guidelines/configurations/default-ai-model-configuration.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.