Sakura Debuts for Edge AI

EdgeCortix launched its first edge-AI inference chip by hardening its DNA IP, delivering low latency and high power efficiency for applications that fit into 5W to 20W.

Bryon Moyer

EdgeCortix has made a strategic shift from selling AI intellectual property (IP) to selling its own edge-AI inference chips for line-powered systems. The new die, dubbed Sakura, started as a test chip, but the company says customer interest convinced it to offer the chip as a product. The chips come mounted on one of two cards: a dual-M.2 (M-key) card and a low-profile PCIe card.

Based in Japan, EdgeCortix has supplemented its $13.5 million in total funding with revenue from FPGAs and ASICs that employ its soft IP. It targets perception—vision, lidar, and related technologies—for transportation, augmented/virtual reality, industry, smart cities, and drones.

Sakura, revealed first at the recent Linley Spring Processor Conference, implements the company’s dynamic neural accelerator (DNA) engine, adding on-chip SRAM, two LPDDR4X ports, and I/O. The chip has no host CPU, so it operates under the control of an external host. Sakura has a maximum performance of 40 TOPS; on ResNet-50, it achieves 0.4ms latency at 4.7W, yielding 533 inferences per second per watt (IPS/W). The company plans to ship samples on a development board in July, with production anticipated in 1Q23.

EdgeCortix announced its DNA architecture as IP last year. Although its primary focus will be on selling chips, it intends to entertain IP business opportunistically. The company may also license Sakura hard IP for chiplets. It’s open sourcing the front end of its Mera compiler so future formats will be available to the tool.