infoq.com/news/2026/11/aws-lambda-gpu-managed-instances AWS Lambda Now Runs GPU Inference with 48 Concurrent Invocations Sharing One Model per GPU
AWS has introduced GPU-backed Lambda functions through Managed Instances, letting container-image workloads run on select EC2 G-series instances and share a single model copy across as many as 48 concurrent invocations per environment. The generally available feature is billed at the EC2 On-Demand rate plus a 9.75% management fee and the standard $0.20 per million requests.