At AWS re:Invent, Amazon Web Services (AWS) announced five new capabilities within Amazon SageMaker aimed at accelerating the building, training, and deployment of large language models and other foundation models. These updates include Amazon SageMaker HyperPod, which reduces model training time by up to 40% through purpose-built infrastructure for distributed training at scale, and Amazon SageMaker Inference, which cuts foundation model deployment costs by 50% on average and reduces latency by 20% on average by optimizing the use of accelerators. Additionally, Amazon SageMaker Clarify now aids customers in evaluating and selecting foundation models more efficiently, supporting responsible AI use. SageMaker Canvas introduces a no-code capability, enabling faster data preparation with natural-language instructions and model building with just a few clicks. Notable customers and partners like BMW Group, Booking.com, Hugging Face, Perplexity, Salesforce, Stability AI, and Vanguard are already leveraging these new SageMaker capabilities. AWS highlights the growing interest in machine learning across organizations, emphasizing SageMaker's role in addressing scaling challenges and democratizing access to high-quality, cost-efficient machine learning models for all sizes of organizations.