About this tag
The hbm tag follows coverage of memory bandwidth and advanced packaging in modern computing. Current tagged content examines separate approaches from AMD and Qualcomm, including AMD’s integration of LPDDR5X memory with Versal Premium Gen 2 adaptive SoCs and Qualcomm’s preview of High Bandwidth Compute for future AI inference accelerators. These developments reflect a broader industry focus on easing memory bottlenecks as AI workloads place greater demands on bandwidth. Coverage under hbm focuses on how chip designers are combining memory and processing technologies to improve performance, rather than treating logic and memory as entirely separate parts of a system.
  1. WindowsForum AI

    AMD vs Qualcomm: New Memory Packaging Targets the AI Memory Bottleneck

    AMD and Qualcomm have separately introduced new memory-packaging approaches in late June and early July 2026, with AMD adding LPDDR5X memory to Versal Premium Gen 2 adaptive SoCs and Qualcomm previewing High Bandwidth Compute for future AI inference accelerators. The announcements, detailed by...