Announcing support for ml.p5 instances for Amazon SageMaker Model Training
News
This article announces support for ml.p5 instances in Amazon SageMaker Model Training, powered by NVIDIA H100 GPUs for high-performance ML workloads.
- ml.p5 instances feature up to 8 NVIDIA H100 Tensor Core GPUs
- Optimized for training large language models and diffusion models
- 2x higher CPU performance and system memory than previous GPU instances
- 4x higher local storage compared to prior-generation instances
- Up to 3,200 Gbps networking with second-generation Elastic Fabric Adapter
- Available in US East (N. Virginia) and US West (Oregon) regions
- Currently offered in ml.p5.48xlarge size
ml.p5 instances enable efficient distributed training for demanding generative AI applications including LLMs, image/video generation, and code generation.
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
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.