> For the complete documentation index, see [llms.txt](https://docs.nexos.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nexos.ai/gateway-api/request-types-dashboard.md).

# Request types dashboard

The Request types dashboard gives you detailed visibility into your Gateway API usage by request type. See what your team is building, how much it costs, and which models are being used most.

### Where to find it

You can access the Request types dashboard from your Gateway platform:

{% stepper %}
{% step %}
Open nexos.ai **Gateway**.
{% endstep %}

{% step %}
In the left sidebar, select **Request types**.
{% endstep %}

{% step %}
You'll see all metrics and usage data.
{% endstep %}
{% endstepper %}

{% hint style="info" %}
**Note**: You can only see request type data for teams, users, and API keys you have access to. Request type metrics respect your permissions.
{% endhint %}

### Filter and refine your view

At the top of the dashboard, use the filters to customize what you see:

* **Time range** - Select a date range to analyze usage over a specific period (e.g., last 7 days, last month).
* **Teams** - Filter to see usage from specific teams only.
* **Users** - Filter to see usage from specific users only.
* **Team API Keys** - Filter by which API key was used to make the requests.

Apply multiple filters at once to get very specific views of your usage.

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2F6uhFNhF55Wg0wrNKTjX6%2Fnexos_filter_buttons_clickable.gif?alt=media&amp;token=8a19a4e8-434d-4bf4-9eb8-6ed2c05adf0f" alt=""><figcaption></figcaption></figure>

### What you can see

The Request types dashboard shows several views of your API usage:

#### Summary cards

At the top, you'll see four key metrics:

* **Largest by spend** – which request type costs the most overall
* **Most requests** – which request type has the highest volume
* **Highest cost per request** – which request type is most expensive per call
* **Most popular agent** – the agent generating the most requests

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2F5pVWujwP34cbHtet1r7E%2Fnexos_request_cards_clickable.gif?alt=media&amp;token=64a47cba-56ce-4a58-bf9f-d101cd0e84c3" alt=""><figcaption></figcaption></figure>

#### Requests by type (pie chart)

A breakdown of all requests by type, showing volume and percentage of total requests.

You can see:

* total requests in the period
* each request type as a colored segment
* the number of requests for each type
* what percentage each type represents

#### Requests by agent (pie chart)

Shows which agents are generating the most requests, helping you understand which AI agents are being used most heavily.

This includes:

* agents detected in your requests
* request count per agent
* percentage of total agent requests

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2F85Z90PqYnRxeEh0OPC0X%2Fnexos_requests_by_agent.gif?alt=media&amp;token=fdc0dfc4-00b9-437d-a868-8b6bf4e3c9ed" alt=""><figcaption></figcaption></figure>

#### Share of requests vs share of spend

A comparison view showing how requests and spending align across request types.

This helps you identify:

* request types that are high-volume but low-cost
* request types that are low-volume but high-cost
* types that are more or less expensive than average

Each row shows:

* Request type
* Share of requests (light bar)
* Share of spend (dark bar)
* Cost (dollar amount)

If the spend bar is longer than the request bar, that type costs more than average per request.

#### Spend mix over time

A stacked area chart showing how your spending breaks down across request types over the selected time period.

This shows:

* 100% stacked share of spend
* changes in spending composition day by day
* which types are growing or shrinking

#### Request types detailed table

A comprehensive table listing each request type with detailed metrics:

* Request type – the name and category
* Requests – total number of requests (with trend indicator)
* Cost per 1K requests – the price per thousand requests
* Input tokens – total tokens sent to the model
* Output tokens – total tokens received from the model
* Top model by spend – which model used most for this type
* Share of spend – percentage of total spending

You can sort by any column and click into a request type for more details.

### Dive deeper into request types

Click on any request type row to see more detailed information:

* request sub-types (if applicable)
* which teams or users are using it most
* which models are preferred for that type
* cost trends over time

### View usage by team

Scroll down to see a Request types by team heatmap. This shows:

* each team as a row
* each request type as a column
* the percentage of that team's requests that are each type
* darker cells indicate a larger share

This helps you understand which teams are using which request types.

### Common use cases

* **Monitor cost drivers –** Use the Share of requests vs share of spend view to identify request types that cost more than you'd expect and optimize accordingly.
* **Track agent adoption –** Use the Requests by agent view to see which agents are in heavy use and which are underutilized.
* **Understand team usage patterns** – Use the Request types by team view to see if usage patterns differ across teams and identify opportunities for standardization.
* **Analyze spending trends** – Use the Spend mix over time chart to see if spending composition is changing over time and what's driving the change.
* **Optimize costs** – Use the Cost per 1K requests column to identify expensive request types and consider optimization opportunities.

### FAQ

**What counts as a request?**\
A request is counted each time an API call is made to the Gateway. Only completions, messages, and responses requests are classified into specific types. Everything else shows as Unclassified.

**Why is my request type showing as Unclassified?**\
Unclassified requests don't fit into standard request type categories, typically health checks, metadata calls, or requests too ambiguous to classify. A small percentage is normal.

**How often is the data updated?**\
Dashboard data updates in real time as requests are processed through the Gateway.

**Can I compare request types across different time periods?**\
Yes. You can change the time range filter to see how usage and spending have changed over time.

**What's the difference between input and output tokens?**\
Input tokens are the tokens in your prompt or request. Output tokens are the tokens in the model's response. The sum of both determines the cost of the request.

**Can I export this data?**\
Yes. Look for an export button on the dashboard to download usage data for further analysis.

<br>

<br>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.nexos.ai/gateway-api/request-types-dashboard.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
