> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cloosphere.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Usage

> Track and analyze token usage and API request count by model/user/team

<Info>Admin › Monitoring › Usage</Info>

The usage dashboard **visualizes AI resource consumption in real time**.

* Analyze token usage and request count across multiple dimensions (model, user, agent, group, team).

<Frame caption="See token, request, and user status at a glance under Admin > Monitoring > Usage">
  <img src="https://mintcdn.com/cloocus/Bh7B8Te7n4QnjOi2/images/monitoring/usage-dashboard.png?fit=max&auto=format&n=Bh7B8Te7n4QnjOi2&q=85&s=37e5a71562c5c8ec52b6ee5f95e37aa3" alt="Usage dashboard" width="1918" height="981" data-path="images/monitoring/usage-dashboard.png" />
</Frame>

***

## Key Metrics

Key metrics appear at the top of the dashboard as cards.

| Metric                 | Description                            |
| ---------------------- | -------------------------------------- |
| **Online**             | Currently connected users (real-time)  |
| **Total Tokens**       | All tokens used in the selected period |
| **Total Requests**     | API call count                         |
| **Active Users**       | Active users in the period             |
| **Active Chats**       | Active chat sessions in the period     |
| **Models Used**        | Number of model types used             |
| **Avg Tokens/Request** | Average token consumption per request  |

<Tip>
  If average tokens/request is abnormally high, excessive system prompts or unnecessary context may be included. Review agent settings.
</Tip>

***

## Usage Trend Chart

Visualize usage over time as a line chart.

<Frame caption="Usage trend chart — daily token line graph">
  <img src="https://mintcdn.com/cloocus/Bh7B8Te7n4QnjOi2/images/monitoring/usage-trend-chart.png?fit=max&auto=format&n=Bh7B8Te7n4QnjOi2&q=85&s=5c341fedfc201312b134f89604085993" alt="Usage trend chart" width="1630" height="280" data-path="images/monitoring/usage-trend-chart.png" />
</Frame>

### Period Selection

| Period     | Description              |
| ---------- | ------------------------ |
| **1d**     | Last 1 day               |
| **7d**     | Last 7 days              |
| **30d**    | Last 30 days             |
| **All**    | Entire period            |
| **Custom** | Pick date range directly |

### Chart Metric

The chart shows **token usage trend** as a line graph.

| Metric     | Description               |
| ---------- | ------------------------- |
| **Tokens** | Total token usage by date |

***

## Multi-dimensional Analysis

Break usage down by model, user, agent, group, and team.

* All charts respond to the top filters (period, model, user, group, team, agent).

<Tabs>
  <Tab title="By Model">
    <Frame>
      <img src="https://mintcdn.com/cloocus/Jjr6o_i720oXQtQO/images/monitoring/usage-by-model.png?fit=max&auto=format&n=Jjr6o_i720oXQtQO&q=85&s=c8cc3b8cc650cc4fd9bb53f30e3c6d5e" alt="Per-model usage donut chart" width="1044" height="640" data-path="images/monitoring/usage-by-model.png" />
    </Frame>

    Check how much each AI model is used, by token share.
  </Tab>

  <Tab title="By User">
    <Frame>
      <img src="https://mintcdn.com/cloocus/Jjr6o_i720oXQtQO/images/monitoring/usage-top-users.png?fit=max&auto=format&n=Jjr6o_i720oXQtQO&q=85&s=b37cad31003c693c46ad6522dd0cd822" alt="Top users bar chart" width="1044" height="640" data-path="images/monitoring/usage-top-users.png" />
    </Frame>

    Ranks users by token usage.
  </Tab>

  <Tab title="By Agent">
    <Frame>
      <img src="https://mintcdn.com/cloocus/Jjr6o_i720oXQtQO/images/monitoring/usage-top-agents.png?fit=max&auto=format&n=Jjr6o_i720oXQtQO&q=85&s=563a8ace6034e4d0d7bfac2d97aefaba" alt="Top agents bar chart" width="1044" height="640" data-path="images/monitoring/usage-top-agents.png" />
    </Frame>

    Ranks workspace agents by token usage.

    <Tip>
      Compare agents to identify inefficient ones and optimize prompts.
    </Tip>
  </Tab>

  <Tab title="By Group">
    <Frame>
      <img src="https://mintcdn.com/cloocus/iRgCRr3H3czOaMhv/images/monitoring/usage-top-groups.png?fit=max&auto=format&n=iRgCRr3H3czOaMhv&q=85&s=e14bfba4d76f1cf01b2c2dab2f6c0691" alt="Top groups bar chart" width="1044" height="640" data-path="images/monitoring/usage-top-groups.png" />
    </Frame>

    Ranks permission groups by token usage. When there is no usage in the selected period, it shows "No group usage data".
  </Tab>

  <Tab title="By Team">
    <Frame>
      <img src="https://mintcdn.com/cloocus/Jjr6o_i720oXQtQO/images/monitoring/usage-top-teams.png?fit=max&auto=format&n=Jjr6o_i720oXQtQO&q=85&s=79fdb9b1c6b478af3b7e3c1dec95852e" alt="Top teams bar chart" width="1044" height="640" data-path="images/monitoring/usage-top-teams.png" />
    </Frame>

    Ranks Teams by token usage.
  </Tab>
</Tabs>

***

## Usage Type Classification

Classify what work tokens were used for.

* The chart shows **only types with usage records in the selected period**.

<Frame caption="Distribution by usage type">
  <img src="https://mintcdn.com/cloocus/Bh7B8Te7n4QnjOi2/images/monitoring/usage-bottom.png?fit=max&auto=format&n=Bh7B8Te7n4QnjOi2&q=85&s=de75391f96f2240674c7f524b2b889b4" alt="Distribution by usage type" width="805" height="338" data-path="images/monitoring/usage-bottom.png" />
</Frame>

<Accordion title="Usage types" icon="list">
  Only types with **actual activity in the period** appear in the chart.

  **User activity (counts toward the daily limit)**

  | Type                          | Description                                 |
  | ----------------------------- | ------------------------------------------- |
  | **chat**                      | Standard chat response generation           |
  | **generation**                | Agent/RAG response generation               |
  | **agent\_state**              | Agent state / next-step decision            |
  | **reasoning**                 | Reasoning step                              |
  | **tool\_call**                | Tool invocation                             |
  | **function\_calling**         | Function call                               |
  | **embedding**                 | Document/search embedding                   |
  | **image\_generation**         | Image generation                            |
  | **code\_gateway**             | Code Gateway execution                      |
  | **moa\_response\_generation** | MoA (Mixture-of-Agents) response generation |

  **Background tasks (not counted toward the limit)**

  | Type                          | Description                        |
  | ----------------------------- | ---------------------------------- |
  | **title\_generation**         | Chat title auto-generation         |
  | **tags\_generation**          | Chat tag auto-generation           |
  | **emoji\_generation**         | Chat emoji auto-generation         |
  | **query\_generation**         | Search query generation            |
  | **image\_prompt\_generation** | Image-generation prompt enrichment |
  | **autocomplete\_generation**  | Input autocomplete suggestion      |
  | **follow\_up\_generation**    | Follow-up question suggestion      |

  **System / monitoring**

  | Type                                              | Description                                  |
  | ------------------------------------------------- | -------------------------------------------- |
  | **guardrail**                                     | Guardrail LLM judgment                       |
  | **schema\_extraction**                            | Structured data (schema) extraction          |
  | **scheduled\_task**                               | Scheduled task execution                     |
  | **trace\_analysis**                               | Trace analysis report generation             |
  | **external\_service**                             | Direct gateway call to an external service   |
  | **auto\_eval:{retrieval, faithfulness, quality}** | Auto-evaluation LLM judgment (per eval type) |
</Accordion>

***

## Filtering

Combine multiple conditions to filter data.

* All filters apply simultaneously.

| Filter     | Description           |
| ---------- | --------------------- |
| **Period** | Start/end date range  |
| **Model**  | Pick a specific model |
| **User**   | Pick a specific user  |
| **Group**  | Pick a specific group |
| **Team**   | Pick a specific team  |
| **Agent**  | Pick a specific agent |

<Note>
  Filter dropdowns only show items with actual usage records.
</Note>

<Note>
  Individual conversation details are available in a separate **Monitoring > Conversation Logs** tab.
</Note>

***

## Related Pages

<Columns cols={3}>
  <Card title="General" icon="gear" href="/en/admin/settings/general">
    Configure usage limits and excess actions
  </Card>

  <Card title="Tracing" icon="chart-line" href="/en/monitoring/tracing">
    Track per-request token usage
  </Card>

  <Card title="Auto-Evaluations" icon="star" href="/en/monitoring/auto-evaluations">
    Review response quality evaluation results
  </Card>
</Columns>