AWS introduces Lambda MicroVMs for isolated execution of user and AI-generated code
News
AWS introduces Lambda MicroVMs, a new serverless compute primitive providing VM-level isolation, near-instant launch and resume speeds, and state preservation for executing user or AI-generated code.
- VM-level isolation for secure multi-tenant code execution without managing virtualization infrastructure
- Near-instant launch speeds and ability to suspend/resume execution for up to 8 hours
- Built on Firecracker virtualization, powering 15+ trillion monthly Lambda invocations
- Create MicroVM images from Dockerfiles and launch with dedicated HTTPS URLs supporting HTTP/2, gRPC, WebSockets
- Available in US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Tokyo), Europe (Ireland)
- Pay for baseline compute resources while running, plus active resource consumption above baseline
Lambda MicroVMs eliminate trade-offs between isolation, speed, and state retention for multi-tenant applications executing user or AI-generated code.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Jun 22
2026
2026
Run isolated sandboxes with full lifecycle control: AWS Lambda introduces MicroVMs
Nov 19
2025
2025
AWS Lambda announces new tenant isolation mode to simplify building tenant-aware applications
Apr 10
2026
2026
Building Memory-Intensive Apps with AWS Lambda Managed Instances
Dec 1
2025
2025
Announcing AWS Lambda Managed Instances, a capability to run functions on your Amazon EC2 instances
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.