Amazon SageMaker Notebook Instances now support Trainium1 and Inferentia 2 based instances

We are pleased to announce general availability of Trainium1 and Inferentia2 based EC2 instances on SageMaker Notebook Instances.

Amazon EC2 Trn1 instances, powered by AWS Trainium chips, and Inf2 instances, powered by AWS Inferentia chips, are purpose-built for high-performance deep learning training and inference, respectively. Trn1 instances offer cost savings over other comparable Amazon EC2 instances for training 100B+ parameter generative AI models like large language models (LLMs) and latent diffusion. Inf2 instances deliver low-cost, high-performance inference for generative AI including LLMs and vision transformers. You can use Trn1 and Inf2 instances across a broad set of applications, such as text summarization, code generation, question answering, image and video generation, recommendation, and fraud detection.

Amazon EC2 Trn1 instances are available for SageMaker Notebook Instances in AWS US East (N. Virginia and Ohio) and US West (Oregon) regions. Amazon EC2 Trn1n instances are available for SageMaker NBI in AWS US East (N. Virginia and Ohio). Amazon EC2 Inf2 instances are available for SageMaker NBI in AWS US West (Oregon), AWS US East (N. Virginia and Ohio), AWS EU (Ireland), AWS EU (Frankfurt), AWS Asia Pacific (Tokyo), AWS Asia Pacific (Sydney), AWS Asia Pacific (Mumbai), AWS EU (London), AWS Asia Pacific (Singapore), AWS EU (Stockholm), AWS EU (Paris), and AWS South America (São Paulo).

Visit developer guide for instructions on setting up and using SageMaker Notebook Instances.