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

# View and Analyse NexLLM API Usage Logs and Statistics

> Track every API request with detailed per-call usage logs, then analyse quota trends, token consumption, and model activity in the NexLLM Dashboard.

NexLLM gives you two complementary tools for understanding how your account uses the API: granular **Usage Logs** that record every individual request, and an interactive **Dashboard** that visualises trends and consumption patterns over time. Together they help you stay on top of costs, debug unexpected usage, and optimise your model choices.

***

## Usage Logs

The **Usage Logs** page provides a full, per-request history of every API call your account has made. Open it by clicking **Logs** in the left sidebar.

Each row in the log table represents one API call and displays the following fields:

| Column                | Description                                               |
| --------------------- | --------------------------------------------------------- |
| **Request Time**      | The date and time the request was received.               |
| **API Key Name**      | The name of the API key used to authenticate the request. |
| **Model Name**        | The model that handled the request.                       |
| **Timing**            | How long the request took to complete.                    |
| **Token Consumption** | The number of input and output tokens used.               |
| **Cost**              | The quota deducted for this request.                      |

### Search and Filter Logs

Use the filter bar above the log table to narrow down your records to exactly what you need.

<Steps>
  <Step title="Set a time range">
    Under **Time Range**, select a start date and an end date to limit results to requests made within that window.
  </Step>

  <Step title="Filter by model (optional)">
    In the **Model** field, type a model name or keyword. The results will include only requests that match your entry.
  </Step>

  <Step title="Filter by group (optional)">
    In the **Group** field, enter the name of the channel group you want to inspect.
  </Step>

  <Step title="Apply the filters">
    Click **Search**. The log table refreshes and displays only the records that match your selected criteria.
  </Step>
</Steps>

***

## Usage Statistics (Dashboard)

The **Dashboard** gives you a bird's-eye view of your API activity through charts and summary metrics. Navigate to it by clicking **Dashboard** in the left sidebar.

### Key Metrics

At the top of the Dashboard you will find five summary statistics for the selected time period:

<CardGroup cols={3}>
  <Card title="Total Requests" icon="arrow-up-right-dots">
    The total number of API calls made in the selected period.
  </Card>

  <Card title="Quota Consumption" icon="circle-dollar-to-slot">
    The total quota deducted across all requests.
  </Card>

  <Card title="Token Usage" icon="coins">
    The combined input and output tokens consumed.
  </Card>

  <Card title="Average RPM" icon="gauge-high">
    The average number of requests per minute over the period.
  </Card>

  <Card title="Average TPM" icon="chart-line">
    The average number of tokens per minute over the period.
  </Card>
</CardGroup>

### Time Range and Preferences

Use the **Filter** controls to choose what data the Dashboard displays:

* **Quick ranges** — Select **1 Day**, **7 Days**, **14 Days**, or **29 Days** for instant presets.
* **Custom range** — Pick any start and end date. The maximum custom duration is **1 month**.

The **Preferences** panel lets you save your preferred defaults — including the default time range, time granularity, consumption chart type, and model call chart type — so the Dashboard always opens with the view you find most useful.

<Tip>
  Use the Quota Distribution chart to pinpoint high-cost time periods within a day or week. Once you spot a spike, cross-reference it with the Usage Logs filtered to that same window to identify which API key and model drove the increase.
</Tip>

***

### Quota Distribution

The **Quota Distribution** section shows how your quota spend breaks down across the selected time range, giving you a clear picture of when and where resources are being consumed.

From this chart you can:

* Review **daily or hourly** quota consumption trends.
* **Compare quota usage across different models** to see which ones cost the most.
* Identify **periods with unusually high consumption** that may warrant investigation.
* Switch between **Bar Chart** and **Area Chart** views depending on how you prefer to visualise the data.
* **Hover over any data point** to see the exact quota value for that time slice and model.

***

### Model Call Analytics

The **Model Call Analytics** section focuses on request volume rather than cost, showing you how frequently each model is called during the selected period.

You can use this section to:

* Monitor your **total model call volume** over time.

* View **call trends** to see whether activity is rising or falling.

* **Compare request activity between models** to understand which ones you rely on most.

* Identify the **most frequently called models** across your account.

* Switch between three dedicated views:

  | View                        | What it shows                                                |
  | --------------------------- | ------------------------------------------------------------ |
  | **Call Trend**              | A time-series chart of call volume over the selected period. |
  | **Call Count Distribution** | A breakdown of total calls by model.                         |
  | **Call Count Ranking**      | Models ranked by total call count, highest first.            |

* **Hover over the chart** to view exact call counts for any point in time.