A California-based AI-chip startup pursuing in-memory computing, an architecture in which the same components handle both memory and computation. By eliminating the need to shuttle data between separate processing and memory units—the so-called "memory wall" that bottlenecks conventional AI inference—the design promises lower energy use and faster inference. d-Matrix expects to release its first widely available chip in 2026.
I don't mind arguing with myself. It's when I lose that it bothers me.