Amazon EC2 P6-B300 instances with NVIDIA Blackwell Ultra GPUs are now available

Amazon EC2 P6-B300 Instances Now Available
AWS has announced the general availability of Amazon EC2 P6-B300 instances, powered by NVIDIA Blackwell Ultra B300 GPUs. These instances offer 8x GPUs with 2.1 TB of high bandwidth GPU memory, 6.4 Tbps EFA networking, 300 Gbps dedicated ENA throughput, and 4 TB of system memory.
Compared to P6-B200 instances, P6-B300 instances provide 2x networking bandwidth, 1.5x GPU memory size, and 1.5x GPU TFLOPS (at FP4, without sparsity), making them ideal for training and deploying large foundation models and language models with advanced techniques. The enhanced networking and larger memory result in faster training times and higher token throughput for AI workloads.
What to do
- Reserve P6-B300 instances through Amazon EC2 Capacity Blocks for ML or Savings Plans.
- Contact your account manager for on-demand reservations.
- Visit Amazon EC2 P6 instances for more information.
Source: AWS release notes
If you need further guidance on AWS, our experts are available at AWS@westloop.io. You may also reach us by submitting the Contact Us form.



