Tools · AWS ML Blog ·
Enhancing enterprise inference on Amazon SageMaker HyperPod with data capture, Hugging Face, NVMe, and Route 53 integration
AWS announced new SageMaker HyperPod inference capabilities: multi-tier data capture for auditing and model improvement, direct deployment from Hugging Face Hub, local NVMe model loading to reduce cold starts, automated Route 53 DNS for custom domains, and pod-level IAM via custom service accounts.