Amazon EC2 P6-B300 Instances Expand to Asia Pacific and South America with Enhanced AI Capabilities

AWS released P6-B300 instances featuring advanced NVIDIA Blackwell Ultra GPUs to additional regions, targeting large-scale AI model training and deployment.

By Elena Parker · Editorial pseudonym · 2026-08-28 · 1 min read

AI-ASSISTED REPORTING · RETROSPECTIVE ARCHIVE

Retrospective edition: 2026-08-28. Source published: 2026-08-28. Added to this archive: 2026-11-03. This briefing was prepared later from the linked announcement; it was not published on this website on the historical edition date.

Amazon EC2 P6-B300 instances became available in the Asia Pacific (Hyderabad) and South America (Sao Paulo) regions. These instances include eight NVIDIA Blackwell Ultra GPUs with 2.1 TB of GPU memory and 4 TB system memory, delivering significant increases in GPU memory and computational throughput compared to P6-B200 models. High bandwidth networking and Elastic Fabric Adapter support enhance distributed training efficiency.

Designed for training trillion-parameter foundation and large language models, the P6-B300 instances improve AI workload speed and throughput by leveraging higher memory capacity and doubled network bandwidth. These offerings are available across multiple global regions, supporting large-scale AI infrastructure expansion.

Source & further reading

Amazon EC2 P6-B300 instances are now available in additional AWS Regions ↗

Source published 2026-08-28. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.