Amazon ECS Managed Instances add support for AWS Trainium and Inferentia
Amazon ECS Managed Instances now supports AWS Trainium and AWS Inferentia, purpose-built AI accelerators for generative AI workloads. This feature simplifies infrastructure management for ECS users by offloading operations to AWS, enhancing performance and reducing costs. Users can now select accelerated instance types and configure resource requirements within their task definitions to allocate accelerators efficiently. This update is available immediately via the AWS Console, ECS MCP Server, or IaC tooling.
- →ECS Managed Instances now supports AWS Trainium and Inferentia accelerators
- →Simplified accelerator allocation for ECS tasks
- →Availability and Getting Started
Features (1) ›
- ECS Managed Instances now supports AWS Trainium and Inferentia accelerators
Amazon ECS Managed Instances can now be provisioned with AWS Trainium and AWS Inferentia, purpose-built AI accelerators for training and inference. This allows users to leverage these accelerators for generative AI workloads with reduced infrastructure management overhead.
Enhancements (1) ›
- Simplified accelerator allocation for ECS tasks
Users can now configure ECS Managed Instances capacity providers to select accelerated instance types, such as Inferentia2 or Trainium1/2. The NEURON_CORE=all configuration ensures optimal performance by allocating all accelerator resources to a single task per instance.
Notes (1) ›
- Availability and Getting Started
This update is available for use with the AWS Console, Amazon ECS MCP Server, or infrastructure-as-code tooling. Users will incur charges for compute management in addition to standard Amazon EC2 costs.
https://aws.amazon.com/about-aws/whats-new/2026/06/amazon-ecs-managed-instances-neuron