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

# Request Logs

> Inspect, filter, and analyze individual API requests, responses, and execution details.

## Overview

The **Request Logs** page provides detailed visibility into every request processed within the selected tenant. It is primarily used for **debugging**, **auditing**, **performance analysis**, and **operational monitoring**.

This page supports real-time inspection, advanced filtering, and historical analysis.

***

## Log Views

### Request Logs

Displays individual request-level logs for model, agent, and API activity.

### Audit Logs

Tracks system and administrative actions such as configuration changes, user actions, and access events.\
(Use Audit Logs for compliance and governance reviews.)

***

## Searching & Filtering

### Search by Request ID

Use the search bar to quickly locate a specific request using its **Request ID**.

### Filters

Select **Filters** to narrow results using available dimensions such as:

* Time range
* Status (success or failure)
* Model
* Team
* Virtual key
* User or agent
* Tags (if applied)

Use **Reset Filters** to clear all applied filters.

***

## Time Range Selection

Use the time selector to control how much historical data is displayed.

Available presets include:

* Last 15 minutes
* Last hour
* Last 4 hours
* Last 24 hours
* Last 7 days
* Custom range

All visible logs and metrics update automatically when the time range changes.

***

## Live Tail & Refresh

### Live Tail

Enable **Live Tail** to stream new request logs in near real time as they occur.

**Common use cases:**

* Observing live traffic during deployments
* Debugging active integrations
* Monitoring agent execution in real time

### Refresh

Manually reload log data when Live Tail is disabled.

***

## Log Table Columns

Each row in the table represents a single request and includes the following fields:

* **Time**\
  Timestamp when the request was processed.

* **Status**\
  Indicates whether the request succeeded or failed.

* **Session ID**\
  Identifier used to group related requests within a session.

* **Request ID**\
  Unique identifier for the request (clickable for details).

* **Cost**\
  Cost associated with the request.

* **Duration (s)**\
  Total execution time in seconds.

* **Team Name**\
  Team under which the request was executed.

* **Key Hash**\
  Obfuscated identifier of the virtual key used.

* **Key Name**\
  Human-readable alias of the virtual key.

* **Model**\
  Model used to process the request.

* **Tokens**\
  Token usage for the request, typically shown as input + output.

* **Internal User**\
  Internal user or system identity that initiated the request.

* **End User**\
  End-user identifier passed by the client (if provided).

* **Tags**\
  Tags associated with the request for tracking and attribution.

***

## Pagination

When large volumes of data are present:

* Results are paginated
* Navigation controls allow moving between pages
* The total number of results and current page are shown

***

## Common Use Cases

* Debug failed or slow requests
* Trace specific user or agent activity
* Validate routing and model selection
* Audit API key usage
* Investigate cost or token anomalies
* Monitor live traffic during rollouts

***

## Best Practices

* Use **Request ID search** for precise debugging
* Combine **filters** with **time ranges** to narrow investigations
* Enable **Live Tail** during active testing or production troubleshooting
* Use **Tags** to simplify log analysis and cost attribution
* Reference **Audit Logs** for administrative or security reviews