Posted On: May 7, 2024

Agents for Amazon Bedrock enable developers to create generative AI-based applications that can complete complex tasks for a wide range of use cases and deliver answers based on company knowledge sources. As agentic applications scale, they require higher input and output model throughput compared to on-demand limits.

Today, we are launching support for Provisioned Throughput with Agents for Amazon Bedrock. With Provisioned Throughput, you can purchase model units for the specific base model. A model unit provides a certain guaranteed throughput, which is measured by the maximum number of input or output tokens processed per minute. You are charged by the hour for each model unit and you have the flexibility to choose between no commitment, and 1-month or 6-month commitment terms.

To learn more about the new capabilities on Bedrock Agents, visit the documentation page.