Workload Usage Dashboard

Updated: July 31, 2026

The Workload Usage Dashboard provides a real-time overview of your organization's AI connections, offering insights into AI services, resources, logs, and token usage. It helps monitor and manage AI interactions across different providers.

Navigation

To access the Workload Usage Dashboard:

  • In the side menu, go to Dashboards.
  • Click on the Workload Usage indicator near the top of the page.

Use the duration selector in the top right of the page to change the period every chart covers.

Connected Services

AI services act as the interface between users and AI models, processing inputs and returning outputs. This section displays:

  • All connected AI services
  • Available AI services that can be connected, and those marked Coming soon

To add new services, click an available service on the dashboard. Learn more here.

Summary

Resources Grouped by Provider

This section organizes AI resources by provider (for example, Anthropic, Amazon, Google Cloud, OpenAI) and displays:

  • Number of models discovered
  • Number of prompts
  • Logs
  • Input and output tokens
  • Total tokens per provider

Logs Grouped by Provider

A chart displaying the number of logs generated by each AI service over the selected time period. Hover over the chart to see a breakdown of logs per provider for each time period.

Total Tokens

A graph showing token usage trends over time. Hover over the graph for a detailed breakdown of token metrics by date.

Log Volume

Log Volume by Log Type

Cloud log volume over time, broken down by the type of log recorded.

Log Volume by Region

The same log volume split by the cloud region the activity came from, which helps surface unexpected regional usage.

Token Usage

Input Tokens by Provider

Input token consumption over the selected period, with a line per provider.

Output Tokens by Provider

Output token consumption over the selected period, with a line per provider.

Total Tokens by Model

Total token consumption attributed to each model, which identifies the models driving the most usage.

Total Tokens by Project

Total token consumption attributed to each project, for chargeback or budget tracking.