---
title: Hero AI Tokens
slug: turbine-user-guide/hero-ai-tokens
docTags: 
createdAt: 2026-07-01T04:52:29.696Z
---

The **Hero AI Tokens** dashboard tracks **token volume** for Hero AI usage across your account. Use it to see input, output, and cache token consumption for playbook automation or user-facing Hero AI features, filter by tenant and model, and review trends over time.

:::hint{type="info"}
**Hero AI Tokens** appears under **Admin Panel > Usage**. You need **Admin Panel** read access and **Usage** read permission. For **credit** consumption, see [Hero AI Credits](docId\:FPBH2J8Xm1TfDvEjnc8KB). For **prompt counts**, see [Hero AI Prompts](docId\:RV7WVN3wx2Uwa_bCHY7uR).

Starting in **Turbine 26.2.0**, the legacy **Hero AI** usage page is removed from **Admin Panel > Usage**.

Token counting is **not retroactive**. This dashboard shows token usage accumulated only **after** the **26.2.0** release.
:::

## Choose Your Path

| I want to…                                  | Start here                                                                          |
| ------------------------------------------- | ----------------------------------------------------------------------------------- |
| Open the tokens dashboard                   | [Open Hero AI Tokens](https://docs.swimlane.com/hero-ai-tokens#open-hero-ai-tokens) |
| See tokens from playbooks or components     | [Playbook Usage Tab](https://docs.swimlane.com/hero-ai-tokens#playbook-usage-tab)   |
| See tokens from companion and user features | [User Prompts Tab](https://docs.swimlane.com/hero-ai-tokens#user-prompts-tab)       |
| Understand token categories                 | [Token Categories](https://docs.swimlane.com/hero-ai-tokens#token-categories)       |

## Open Hero AI Tokens

1. Open **Admin Panel**.
2. Go to **Usage**.
3. Click **Hero AI Tokens**.

## Token Categories

Hero AI token usage is grouped into four categories:

| Category               | Description                                          |
| ---------------------- | ---------------------------------------------------- |
| **Input Tokens**       | Tokens sent to the model in the request              |
| **Output Tokens**      | Tokens returned by the model in the response         |
| **Cache Read Tokens**  | Tokens read from prompt cache (when caching applies) |
| **Cache Write Tokens** | Tokens written to prompt cache                       |

Summary cards, charts, and **Total Tokens** values on each tab reflect these categories. Hover **Total Tokens** in the usage table to see the breakdown for each row.

## Playbook Usage Tab

The **Playbook Usage** tab shows token volume tied to **playbook and component automation**.

:::hint{type="info"}
**Test** runs from an unsaved **Hero AI Native Action** may show placeholder **Action** and **Name** values (for example **Unsaved Playbook** and **Unsaved Flow**). See [Hero AI Credits](docId\:FPBH2J8Xm1TfDvEjnc8KB) in [Hero AI Credits](docId\:FPBH2J8Xm1TfDvEjnc8KB).
:::

### Filters

| Filter         | Description                                                          |
| -------------- | -------------------------------------------------------------------- |
| **Date Range** | Today, Last 7 days, Last 30 days, Last 90 days, or Custom date range |
| **Tenant**     | One or more tenants (multi-tenant accounts)                          |
| **Type**       | **Playbook** or **Component**                                        |
| **Name**       | Filter by playbook or component name                                 |
| **Model Type** | Filter by model tier used for the usage row                          |

### Summary Cards and Chart

| Element                      | Description                                                                           |
| ---------------------------- | ------------------------------------------------------------------------------------- |
| **Total Input Tokens**       | Input tokens for the filtered selection                                               |
| **Total Output Tokens**      | Output tokens for the filtered selection                                              |
| **Total Cache Read Tokens**  | Cache read tokens for the filtered selection (shows **--** when no cache read data)   |
| **Total Cache Write Tokens** | Cache write tokens for the filtered selection (shows **--** when no cache write data) |
| **Hero AI TOKENS** chart     | Stacked bar chart of token usage over time by category                                |

### Usage Table

| Column           | Description                                                                                                                                                                                                                                                                  |
| ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Tenant**       | Tenant where usage occurred                                                                                                                                                                                                                                                  |
| **Type**         | **Playbook** or **Component**                                                                                                                                                                                                                                                |
| **Action**       | Hero AI action identifier. When you **Test** a **Hero AI Native Action** before the playbook is saved, this may show an auto-generated placeholder (for example `heroAI_ahecv`) instead of your action title.                                                                |
| **Name**         | Playbook and flow name, or component name. Links open the canvas in a new tab when the item still exists. **Unsaved Playbook** and **Unsaved Flow** appear when usage came from **Test** on an unsaved playbook. **Deleted** labels appear when the source no longer exists. |
| **Model Type**   | Model tier associated with the usage                                                                                                                                                                                                                                         |
| **Total Runs**   | Number of runs in the filtered period                                                                                                                                                                                                                                        |
| **Total Tokens** | Combined token count. Hover for input, output, cache read, and cache write breakdown.                                                                                                                                                                                        |

## User Prompts Tab

The **User Prompts** tab shows token volume from **Hero AI Companion** and related prompt-based features.

### Filters

| Filter         | Description                                                          |
| -------------- | -------------------------------------------------------------------- |
| **Date Range** | Today, Last 7 days, Last 30 days, Last 90 days, or Custom date range |
| **Tenant**     | One or more tenants                                                  |
| **User**       | One or more users                                                    |

### Summary Cards and Chart

The **User Prompts** tab shows the same four token summary cards and a stacked **Hero AI TOKENS** chart as the playbook tab, scoped to user prompt usage.

### Usage Table

| Column           | Description                                                                           |
| ---------------- | ------------------------------------------------------------------------------------- |
| **Tenant**       | Tenant where prompts were sent                                                        |
| **User**         | User who sent prompts                                                                 |
| **Prompt count** | Number of prompts in the period                                                       |
| **Total Tokens** | Combined token count. Hover for input, output, cache read, and cache write breakdown. |

## See Also

- [Hero AI Credits](docId\:FPBH2J8Xm1TfDvEjnc8KB) — credit consumption and contract usage caps
- [Hero AI Prompts](docId\:RV7WVN3wx2Uwa_bCHY7uR) — prompt counts
- [Hero AI Models](docId\:C1LdFGRmok-7Z2Nei_28b) — default models and usage reporting overview
- [Usage Dashboard](docId\:Q2iM3wGt2UNLVGV820OTU) — other usage dashboards under **Admin Panel > Usage**
