> For the complete documentation index, see [llms.txt](https://imbrace.gitbook.io/imbrace-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://imbrace.gitbook.io/imbrace-documentation/governance/token-usage.md).

# Token Usage

**Token Usage** analyzes AI token consumption, estimated cost, and usage concentration by model and department. Use it to identify which AI agents and models consume the most tokens and to monitor AI costs over time.

<figure><img src="https://3400071099-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FypCKq44Psaly6khgrY7x%2Fuploads%2FSWC93x8DBLX5gIvSYug6%2Fimage.png?alt=media&amp;token=b42a554e-4207-4362-a4c4-100b383eec71" alt=""><figcaption></figcaption></figure>

### **Viewing Token Usage**

1. Click the **Token Usage** tab.
2. Review the summary cards at the top of the page. The summary card includes:

<table data-header-hidden><thead><tr><th width="209.39996337890625"></th><th></th></tr></thead><tbody><tr><td><strong>Total Tokens (Last 30d)</strong></td><td>The total number of tokens consumed in the last 30 days.</td></tr><tr><td><strong>Avg. Tokens per Day</strong></td><td>The average number of tokens consumed per day.</td></tr><tr><td><strong>Total Est. Cost</strong></td><td>The estimated total cost of token usage in USD.</td></tr></tbody></table>

3. Review the charts:

* **By Department:** A donut chart showing each department's share of token usage as a percentage. Usage not assigned to a department is shown as **Uncategorized**.
* **By Model:** A bar chart showing the number of tokens (in millions) consumed by each LLM model. The token count is displayed above each bar. Models that have not consumed tokens show **0**.

<figure><img src="https://3400071099-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FypCKq44Psaly6khgrY7x%2Fuploads%2FGzYoVC2xLuxxmUBihA9L%2Fimage.png?alt=media&amp;token=7af2a875-4415-420c-bd7f-c5ef86aa21d4" alt=""><figcaption></figcaption></figure>

4. **Review usage by AI agent**

The table at the bottom of the page lists token usage for each AI Agent. For each AI Agent, the following information is available:

* **AI Agent:** The name of the AI Agent. Agents that run as part of an orchestrated (multi-agent) setup are prefixed with **\[Orchestrated]**.
* **Department:** The department the AI Agent is assigned to.
* **Model:** The LLM model used by the AI Agent, for example, **qwen3.5-27b** or **gpt-4o**. **"default"** indicates the AI Agent uses the organization's default model.
* **Tokens:** The number of tokens consumed by the AI Agent.
* **Est. Cost (USD):** The estimated cost of the tokens consumed.
* **Avg. Cost per 1K Tokens:** The average estimated cost per 1,000 tokens.

<figure><img src="https://3400071099-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FypCKq44Psaly6khgrY7x%2Fuploads%2FDIWq6217ByT2Hj7MkXG0%2Fimage.png?alt=media&amp;token=4a014727-7a3d-4ac6-8411-74fe8cf68ab7" alt=""><figcaption></figcaption></figure>

Use the page numbers or arrow controls at the bottom of the table to move between pages.
