Amazon SageMaker AI announces availability of P5e and G6e instances for Inference
News
Amazon SageMaker has announced the general availability of two new AI inference optimized instances:
- P5e instances powered by NVIDIA H200 Tensor Core GPUs
- G6e instances powered by NVIDIA L40S Tensor Core GPUs
Key features of these instances include:
- P5e (ml.p5e.48xlarge) offers 1128 GB GPU memory, 30 TB NVMe SSD storage, and 192 vCPUs
- Ideal for large language models, multi-modal foundation models, and generative AI applications
- G6e instances provide up to 2.5x better performance compared to g5 instances
- Supports LLMs up to 13B parameters and diffusion models for various media generation
The instances are currently available in US East (Ohio) and US West (Oregon) regions, with pricing details available on AWS SageMaker's website.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Nov 22
2024
2024
Amazon SageMaker Inference now supports G6e instances
Dec 6
2024
2024
Amazon SageMaker introduces new capabilities to accelerate scaling of Generative AI Inference
May 4
2026
2026
Amazon SageMaker AI Now Supports Capacity-Aware Inference with Automatic Instance Fallback
Apr 20
2026
2026
Accelerate Generative AI Inference on Amazon SageMaker AI with G7e Instances
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.