Overview of billing for agents powered by the GitHub Copilot harness

Note

Features in this article are powered by the GitHub Copilot harness, which uses Copilot Credits for usage-based billing. Learn more about Copilot Credits in Overview of billing for agents powered by the GitHub Copilot harness.

Usage-based billing applies to using, building, testing, and evaluating agents. These actions might consume Copilot Credits.

Agents powered by the GitHub Copilot harness use a new, usage-based billing structure. This article describes what's changed and how Copilot credits are measured for these agents.

What's new

  • Usage-based billing: All usage is billed based on consumption and measured in Copilot credits. Creating and using agents consumes credits based on usage and rate of consumption.
  • Credits cover the full experience: Copilot credits are charged for large language model (LLM) tokens, tools (including knowledge and MCPs), and the harness itself. Any experience that uses one of these consumes credits.
  • Billing starts when you start building: Unlike the standard harness, which starts billing after publish, the GitHub Copilot harness charges credits from the moment you start building. Experiences such as creating an automated solution with natural language, previewing and testing the agent, and generating and creating agent evaluations all consume credits.
  • Scope: This billing structure applies only to agents, workflows, and apps powered by the GitHub Copilot harness.

How Copilot credits are purchased and managed

Copilot credits are managed through the Power Platform admin center. Administrators can allocate credits to environments and monitor usage to ensure that agents have the necessary resources. Learn more about purchasing and managing Copilot credits in Purchase and manage Copilot credits.

How Copilot credits are measured

Copilot credits are the unit that measures agent usage. You calculate the total cost based on the sum of the Copilot credits your organization uses. The number of Copilot credits an agent consumes depends on the design of the agent, how often customers interact with it, and the features they use.

Credits charged for each task vary, but you can expect token usage to fall within the ranges shown in the following diagram, depending on task complexity.

Diagram showing task complexity tiers and their estimated Copilot Credit consumption ranges.

View consumed credits

You can view the Copilot credits an agent consumes on its Monitor page in Copilot Studio. Learn more in monitor an agent.

To view credit consumption across an environment or your tenant, go to Licensing > Copilot Studio in the Power Platform admin center.

To understand how Copilot credits consumption is enforced, learn more in the Enforcement policy.

  • Last updated on