10x the Power Efficiency of the Latest GPUs: Inside the Next-Generation AI Chip Developed by SK hynix and TetraMem
In-memory computing, which performs calculations directly within memory, is drawing attention as a way to improve AI power efficiency. A joint US-Korean research team has developed a proprietary memristor SoC with zigzag wiring to eliminate the inefficiency of operations widely used in lightweight AI models, demonstrating extremely high power efficiency that surpasses existing GPUs.