# AI Messages Costs

Here's the updated documentation with clarifications about token-based pricing:

***

### AI Message Costs

The cost of each AI message sent depends on which AI model is being used and will reduce your agency's wallet balance accordingly. **Pricing is based on the total number of input and output tokens used per message.** The costs shown below represent the average cost per message response for typical usage.

Depending on your plan and what AI model you use determines your cost per AI message sent.

#### Average Pricing per AI Chat Message

For agencies your average cost per AI message sent will be:

**OpenAI Models:**

* GPT-4o (2024-11-20): \~$0.0109 /avg message
* GPT-4o-mini: \~$0.0007 /avg message
* GPT-4O (2024-08-06): \~$0.0109 /avg message
* GPT-4.1: \~$0.0088 /avg message
* GPT-4.1-mini: \~$0.0017 /avg message
* GPT-4.1-nano: \~$0.0004 /avg message
* GPT-5: \~$0.0055 /avg message
* GPT-5-mini: \~$0.0011 /avg message
* GPT-5-nano: \~$0.0002 /avg message

**Anthropic (Claude) Models:**

* Claude-3.7 Sonnet: \~$0.0131 /avg message
* Claude-3.5 Haiku: \~$0.0011 /avg message

**xAI Models:**

* Grok-2-1212: \~$0.0088 /avg message

**Note:** Actual costs may vary based on the length and complexity of each conversation. Longer prompts and detailed responses will use more tokens and therefore cost more, while shorter interactions will cost less.

For example, if you use GPT-5-mini, each message response will reduce your wallet balance by approximately $0.0011, though the exact amount depends on the tokens used in that specific interaction.

#### Bring Your Own API Key

For subscribers on the **Full SaaS Mode plan** and **Agency plan**, you can bring your own API key and avoid paying Stammer.ai's cost per message entirely. This will set it up so you are paying the AI provider (OpenAI, Anthropic, or xAI) directly for your AI usage based on their token pricing.

***

**Cost-saving tip:** The most cost-effective models are GPT-5-nano (\~$0.0002), GPT-4.1-nano (\~$0.0004), and GPT-4o-mini (\~$0.0007), while the most capable models include GPT-5 (\~$0.0055) and Claude-3.7 Sonnet (\~$0.0131).

{% content-ref url="/pages/ReiaIq4eiP67bkWLrAha" %}
[Vault](/stammer.ai-docs/chat-ai-agents/vault.md)
{% endcontent-ref %}


---

# 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.stammer.ai/stammer.ai-docs/account-management/agency-wallet/ai-messages-costs.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.