🧔♂️ A friendly human may check it before it goes live. More news here
AWS Lambda adds GPU support for serverless inference
Amazon Web Services (AWS), the cloud computing arm of Amazon, said AWS Lambda now supports GPU-backed functions on Amazon Elastic Compute Cloud (EC2) G-series instances through Lambda Managed Instances.
The update lets developers run inference, media processing, rendering, and scientific computing workloads in container images without managing servers.
AWS said the feature targets workloads that need GPU acceleration but are often built on self-managed infrastructure.
The company added that it is optimized for inference rather than model training, and remains compatible with existing Lambda event sources and orchestration tools.
The feature is generally available. Pricing combines the underlying EC2 instance cost, a 9.75% management fee for GPU instances, and Lambda’s standard request charge.
The launch comes as cloud providers expand serverless GPU options.
Google Cloud said its Cloud Run GPU offering is generally available, and its documentation shows managed container services can use GPUs.
🔗 Source: AWS Blog




