AWS Lambda extends on-demand timeout to 30 minutes
AWS Lambda now supports a 30-minute timeout for on-demand Event Functions, doubling the earlier 15-minute limit. The change targets long workloads such as AI inference and media transcoding. Standard Lambda pricing applies.
By JetherVerse News

AWS Lambda now supports a 30-minute timeout for on-demand Event Functions, doubling the previous 15-minute limit, according to an AWS announcement on November 19, 2026. The change covers synchronous, asynchronous and event source mapping invocations.
On Lambda Managed Instances, synchronous invocations also reach 30 minutes. Asynchronous and event source mapping invocations there keep the 90-minute timeout AWS launched in September 2026. Event Functions run on demand by default and scale automatically for spiky traffic. For steady traffic, teams run them on Lambda Managed Instances, which uses Amazon EC2 instances while keeping the operational simplicity of Lambda.
AWS names AI inference, media transcoding and financial calculations as workloads helped by the change. These needed architectural workarounds when the limit stood at 15 minutes. The 30-minute limit applies with Lambda durable functions and with Step Functions Standard workflows. On-demand Event Functions get the longer timeout in all commercial AWS Regions, and synchronous calls on Managed Instances get the same in Regions where Managed Instances are offered. Teams set the timeout through the Lambda console, the AWS CLI, SDKs, CloudFormation, CDK or SAM. Standard Lambda pricing applies. AWS adds coding assistants such as Kiro, Claude Code and Codex to the list of tools working with the Agent Toolkit for AWS. Customers use Event Functions to build event-driven processors, API backends and data pipelines.
Longer runs mean fewer workarounds such as splitting jobs into chunks or moving work to container services. Lambda bills for duration, so a function running 30 minutes costs more than one finishing in five. Long timeouts also hide slow code, because a stuck task now burns twice as much time before failing.
Developers should review jobs split to fit the 15-minute cap and decide whether a single longer function is simpler. Set alarms on duration before raising any timeout, and keep retry rules tight so a failing job does not run for 30 minutes several times over.
Read the full report at Amazon Web Services.
Source: Amazon Web Services
Get the news in your inbox
The tech stories that matter from JetherVerse News, with the source on every one.

