AWS expands Lambda MicroVMs to four more regions

Regional availability reaches 14 AWS Regions

AWS announced expanded Lambda MicroVMs availability on October 22, 2026, adding Asia Pacific (Hong Kong), Asia Pacific (Malaysia), Asia Pacific (Thailand), and Europe (Spain). The additions bring the total to 14 AWS Regions, according to the AWS release notice.

AWS positions the expansion as a way to run environments closer to end users and help meet latency and data residency requirements.

Dockerfile-based environments with preserved state

AWS describes Lambda MicroVMs as a serverless compute primitive for executing user- or AI-generated code, with VM-level isolation, state preservation, and what it calls “near-instant” launch and resume speeds. Its stated use cases include AI coding assistants, interactive development environments, and vulnerability scanners. These are vendor-described capabilities; the announcement provides no latency benchmarks.

According to AWS, developers create a MicroVM image from a Dockerfile and launch MicroVMs from that image. They can give each user or job a separate environment with a dedicated HTTPS URL supporting HTTP/2, gRPC, and WebSockets. AWS lists the Lambda console, AWS CloudFormation, AWS Cloud Development Kit, and agentic development tools using Agent Toolkit for AWS as ways to use the service.

Evaluate the whole deployment, not just the region

For an interactive tool, use the additional regions as candidates for end-to-end latency testing. Measure the complete request path, including calls to storage and downstream services, rather than selecting a region solely by its proximity to users.

For residency-sensitive deployments, map where application data, logs, and downstream requests are processed or stored before choosing a region. For applications relying on preserved state, define retention, cleanup, and recovery requirements, then verify the service’s documented behavior against them. The regional announcement establishes availability; it should not replace workload-specific performance and operational evaluation.