Speed up your cluster procurement time with Amazon SageMaker HyperPod training plans

AWS introduces SageMaker HyperPod training plans, a new feature designed to help organizations quickly and efficiently procure compute resources for large language model (LLM) training and fine-tuning.

• Addresses challenges of securing high-performance GPU compute capacity for AI model training
• Provides predictable access to accelerated compute resources like P4d, P5, and trn2 instances
• Offers two ways to create training plans: via SageMaker console or AWS CLI
• Can be used with both SageMaker training jobs and SageMaker HyperPod clusters
• Allows organizations to search, reserve, and manage compute capacity with simple interfaces

The solution helps organizations overcome compute resource constraints, reduce training cluster procurement wait times, and accelerate AI initiatives by providing flexible, scalable access to high-performance computing resources.