AgentCore runtime instances are now generally available
AWS has introduced runtime instances for Amazon Bedrock AgentCore, enabling users to run AI agents on their own EC2 instances while AWS manages infrastructure tasks such as provisioning and scaling.
The useful question is what changes for users, developers or buyers, and whether the announcement stays industry context or becomes something people can actually use.
Amazon Web Services announced runtime instances for Amazon Bedrock AgentCore, allowing teams to deploy and operate AI agents on their own Amazon EC2 instances without handling infrastructure management. This new feature provides purpose-built infrastructure for secure, scalable agent operations. Runtime instances complement the existing microVM-based option, offering access to a broader range of EC2 instance types tailored for sustained, resource-intensive, or specialized workloads. AWS handles provisioning, patching, scaling, and lifecycle management, reducing operational overhead for users.
Teams can create a capacity provider via the AWS Management Console, CLI, SDKs, or APIs to specify the EC2 instance types required for their agents, including GPU-accelerated, memory-optimized, and compute-optimized families. Runtime instances support long-running agent sessions of up to 14 days, while the default serverless, microVM-based runtime remains available for shorter sessions of up to 8 hours. Users can select the appropriate compute resources for each agent or mix instance types without altering deployment or invocation methods.
The new runtime instances are available in nine AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Mumbai), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Tokyo), Europe (Frankfurt), and Europe (Ireland). AWS charges for the management of provisioned compute in addition to standard Amazon EC2 costs, ensuring transparent pricing for compute resources used by agents.
To begin using runtime instances, users can refer to the AWS News Blog or AgentCore documentation for guidance. Pricing details are available on the AgentCore pricing page, providing clarity on costs associated with the new feature. This update expands deployment flexibility for AI agents while maintaining AWS’s managed infrastructure approach.