Amazon EC2 G6 instances are now available in the AWS GovCloud (US-East) region for Amazon SageMaker AI inference. These instances leverage NVIDIA L4 Tensor Core GPUs, each equipped with 24 GB of memory, and AMD EPYC processors. G6 instances offer significantly improved deep learning inference performance, up to double that of G4dn instances. This expansion allows government agencies in GovCloud to deploy inference endpoints for generative AI. Such workloads include language models, image generation, and computer vision tasks. These deployments will adhere to strict compliance and data residency regulations. G6 instances provide excellent price-performance for production inference needs within 24 GB of GPU memory. They are now accessible in AWS GovCloud (US-East) in addition to other regions. Interested users can find detailed pricing information on the AWS pricing page. This availability enhances AI capabilities for government entities.