Amazon SageMaker Serverless Inference is now generally available

Amazon has announced the general availability of Amazon SageMaker Serverless Inference across 21 AWS Regions, offering a simplified way to deploy machine learning models for inference.

• Allows deployment of ML models without configuring infrastructure
• Automatically provisions, scales, and manages compute capacity
• Pay only for compute capacity used, billed by millisecond
• Ideal for applications with intermittent or unpredictable traffic
• New features include SageMaker Python SDK support and increased concurrent invocation limit to 200
• Can be created via AWS console, SDK, CloudFormation, or CLI

The service enables machine learning inference without server management, providing a flexible and cost-effective solution for deploying ML models.