ainewsblitz.com

Breaking

Nvidia Considers Cutting High-Bandwidth Memory in Rubin Ultra GPU Amid HBM Shortage

Nvidia Considers Cutting High-Bandwidth Memory in Rubin Ultra GPU Amid HBM Shortage
  • Infra & Chips
  • Foundation Models

Nvidia is weighing whether to equip its next-generation Rubin Ultra accelerator with substantially less high-bandwidth memory than originally planned, a response to a tightening supply of advanced HBM chips that could reshape the flagship AI processor's specifications. The company is testing multiple prototype configurations, some validated at sharply reduced capacities in the range of 192GB to 256GB, according to reporting from The Information. Final specifications remain unsettled.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,915 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year