PC Watch has published an in-depth feature on LENZO and our CGLA (Coarse-Grained Linear Array) architecture. Rather than positioning LENZO as simply “challenging GPUs,” the article focuses on our fundamentally different approach to acceleration — reducing memory movement, eliminating repeated fetch/decode overhead, and maximizing compute efficiency through linear dataflow design. CGLA is built for structured, repeatable workloads where traditional architectures waste energy moving data instead of computing. The piece also explores our FPGA implementation and future roadmap.
Read the full article here (Japanese Only)