Amazon SageMaker AI in 2025, a year in review part 1: Flexible Training Plans and improvements to price performance for inference workloads

In 2025, Amazon SageMaker AI saw dramatic improvements to core infrastructure offerings along four dimensions: capacity, price performance, observability, and usability. In this series of posts, we discuss these various improvements and their benefits. In Part 1, we discuss capacity improvements with the launch of Flexible Training Plans. We also describe improvements to price performance for inference workloads. In Part 2, we discuss enhancements made to observability, model customization, and model hosting.

Flexible Training Plans for SageMaker

SageMaker AI Training Plans now support inference endpoints, extending a powerful capacity reservation capability originally designed for training workloads to address the critical challenge of GPU availability for inference deployments. Deploying large language models (LLMs) for inference requires reliable GPU capacity, especially during critical evaluation periods, limited-duration production testing, or predictable burst workloads. Capacity constraints can delay deployments and impact application performance, particularly during peak hours when on-demand capacity becomes unpredictable. Training Plans ...