Amazon Web Services launches P6e-GB200 UltraServers with NVIDIA Grace Blackwell chips
AWS just rolled out the P6e-GB200 UltraServers, powered by NVIDIA Grace Blackwell Superchips. These servers are built for training and deploying massive AI models, hitting up to 360 petaflops of FP8 compute and packing 13.4 TB of HBM3e GPU memory.
The P6e-GB200 sports 72 NVIDIA Blackwell GPUs linked by 5th-gen NVLink, working as a single unit. Network speeds reach 28.8 Tbps via the latest Elastic Fabric Adapter (EFAv4). It’s designed for trillion-parameter AI models that need huge compute power and memory. 
AWS also announced P6-B200 instances, an 8-GPU setup with Intel Xeon processors. These are aimed at medium to large AI workloads and offer easier migration from previous instances. P6-B200 packs 1.4 TB GPU memory and up to 3.2 Tbps networking.
David Brown, VP of AWS Compute and ML Services, shared details on the gear’s security and stability:
“When customers tell me why they choose to run their GPU workloads on AWS, one crucial point comes up consistently: they highly value our focus on instance security and stability in the cloud. The specialized hardware, software, and firmware of the AWS Nitro System are designed to enforce restrictions so that nobody, including anyone in AWS, can access your sensitive AI workloads and data. Beyond security, the Nitro System fundamentally changes how we maintain and optimize infrastructure. The Nitro System, which handles networking, storage, and other I/O functions, makes it possible to deploy firmware updates, bug fixes, and optimizations while it remains operational. This ability to update without system downtime, which we call live update, is crucial in today’s AI landscape, where any interruption significantly impacts production timelines. P6e-GB200 and P6-B200 both feature the sixth generation of the Nitro System, but these security and stability benefits aren’t new—our innovative Nitro architecture has been protecting and optimizing Amazon Elastic Compute Cloud (Amazon EC2) workloads since 2017.”
The P6e-GB200 UltraServers run on liquid cooling, enabling higher compute density and efficiency. The P6-B200 sticks to proven air cooling.
AWS supports these new GPU instances with Amazon SageMaker HyperPod for managed cluster operations, Amazon EKS for Kubernetes workloads, and NVIDIA DGX Cloud for a complete AI platform.
Watch AWS’s launch video here.
This launch lifts the bar for AI infrastructure at AWS. Expect faster training, scalable inference, and better resource efficiency for next-gen AI workloads.
