Devices · 1h ago
Qualcomm unveils HBC near-memory AI accelerators, claims 6x bandwidth-per-watt over HBM
Qualcomm announced its HBC near-memory AI architecture with AI250 and AI350 accelerators, claiming 6x higher bandwidth-per-watt versus HBM and 200x capacity compared to on-chip SRAM. The design aims to overcome the memory wall limiting AI performance. The accelerators target data center and edge AI workloads.
Meridian48 take
Qualcomm's near-memory approach could challenge HBM dominance, but real-world performance and adoption remain unproven.
Read the full reporting
Qualcomm reveals HBC near-memory AI architecture, AI250 and AI350 accelerators — touts 6x higher bandwidth-per-watt compared to HBM, 200x capacity compared to on-chip SRAM →
Tom's Hardware
ai-acceleratorsnear-memory-computing