AWS Launches P4de Instances in Tokyo with Enhanced GPU Memory and Training Efficiency

Amazon EC2 P4de instances, featuring 8 NVIDIA A100 GPUs with 80GB memory each, offer improved performance and cost savings on SageMaker notebooks.

By Marcus Lane · Editorial pseudonym · 2026-05-27 · 1 min read

AI-ASSISTED REPORTING · RETROSPECTIVE ARCHIVE

Retrospective edition: 2026-05-27. Source published: 2026-05-27. 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.

The P4de instance series doubles GPU memory compared to earlier P4d instances, totaling 640GB GPU memory per instance. This hardware upgrade enables up to 60% better machine learning training performance and 20% lower costs, particularly benefiting workloads requiring large high-resolution data training.

Availability in Asia Pacific (Tokyo) SageMaker notebooks allows developers to reduce model training times and accelerate time to market for AI applications. AWS supplies developer guides for integrating these instances into JupyterLab and CodeEditor within SageMaker Studio environments.

Source & further reading

Announcing Region Expansion of P4de instances on SageMaker Notebook Instances ↗

Source published 2026-05-27. 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.