In this article

1. Message scenarios
2. Overage
3. Frequently asked questions
4. Set up pay-as-you-go consumptive meter
5. View message consumption

This article describes the billing models for messages available with a Copilot Studio license. Prepaid messages are available for purchase since December 1, 2023. Pay-as-you-go is available for purchase since December 1, 2024.

Messages are the unit that measures agent usage. The total cost is calculated based on the sum of the messages used by your organization. The number of messages consumed by an agent depends on the complexity of the agent, how often customers interact with it, and the features they use.

The purchase of a Copilot Studio license includes a specific number of billed messages. This capacity is pooled across the entire tenant. However, it must be assigned to an environment to enable Copilot Studio features for agents in that environment.

Note

As of February 1, 2025, we are introducing autonomous actions billing which charges for each generatively orchestrated action, including triggers, topics, Power Platform connectors, and Power Automate flows. You can see these actions on the activity map when you test your agent in Copilot Studio. Knowledge uses a separate consumption rate through generative messages and tenant Microsoft Graph grounding for message rates. Each autonomous action incurs a charge of 25 messages against your capacity packs, or counts 25 messages against your Copilot Studio pay-as-you-go meter.

Allocation happens in the Power Platform admin center when you select Billing > Licenses > Copilot Studio, and then select Manage Capacity. For more information, see Capacity add-ons, the Power Platform documentation.

Message scenarios

The following table illustrates the types of messages managed by the Microsoft 365 Copilot subscription.

Note

When a Copilot Studio subscription (message pack or meter) is set up for Microsoft 365 Copilot Chat, usage of agents by users not licensed for Microsoft 365 Copilot will consume Copilot Studio messages. Most of these conversations will consume 32 messages per interaction or query (30 messages for tenant Microsoft Graph grounding and 2 for generative answers).

• Classic answers: These events are predefined responses that are manually authored by agent makers. They're static and don't change unless manually updated. They're typically used where precise and controlled responses are the only ones we want the agent to generate.
• Generative answers: These events are dynamically generated using AI models, such as Generative Pretrained Transformers (GPTs). They can adapt and change based on the context and the knowledge sources they're connected to. They're useful for handling a wide range of topics and providing more flexible and natural interactions.
• Tenant Microsoft Graph grounding for messages: These events provide higher quality grounding for your agents using retrieval-augmented generation (RAG) over your tenant-wide Microsoft Graph, including external data synced into Microsoft Graph through connectors. This results in more relevant and improved responses and ensures that the grounding information is up to date. This capability is optional, and you can turn it on or off for each agent.