Chips & Compute·4 min readPreferred Networks Targets Faster AI Inference with MN-Core L2000UUnite.AINov 16