SageMaker Notebook Instances Add Support for G6e EC2 Instances
Amazon SageMaker notebook instances now support EC2 G6e instances, featuring NVIDIA L40s GPUs and AMD EPYC processors. This enhancement offers up to 2.5x better performance than G5 instances, enabling faster interactive model training and deployment for generative AI, LLMs, and diffusion models. G6e instances are available in select AWS regions and are suitable for users working with large models.
- →General availability of EC2 G6e instances on SageMaker notebook instances
- →Regional availability and use cases for G6e instances
Features (1) ›
- General availability of EC2 G6e instances on SageMaker notebook instances
Amazon SageMaker notebook instances now support EC2 G6e instances, powered by NVIDIA L40s Tensor Core GPUs and AMD EPYC processors. These instances offer up to 2.5x better performance compared to G5 instances and are suitable for interactive model training and deployment use cases, including generative AI, LLMs, and diffusion models.
Notes (1) ›
- Regional availability and use cases for G6e instances
G6e instances on SageMaker notebook instances are available in specific AWS regions and can be used for deploying large language models up to 13B parameters and diffusion models. This allows for interactive testing of model deployments and advanced training tasks.
https://aws.amazon.com/about-aws/whats-new/2026/03/g6e-new-launch-sagemaker-notebook-instances/