LUBIS EDA
LUBIS EDA Launches FormalOS: Infrastructure for Systematic, Scalable Formal Verification
EdgeAI Inc. a pioneering South Korean AI semiconductor startup specializing in ultra-low-power neuromorphic hardware, and Andes Technology (TWSE: 6533), a leading global supplier of high-performance, low-power 32/64-bit RISC-V processor intellectual property (IP) and a Founding Premier Member of RISC-V International, today announced the signing of a strategic IP licensing agreement.
Under this agreement, EdgeAI will integrate Andes’ advanced RISC-V CPU cores into its upcoming neuromorphic system-on-chip (SoC) architecture. By pairing Andes’ highly efficient computational pipeline with EdgeAI’s event-driven neuromorphic accelerators, the partnership aims to deliver a breakthrough in energy efficiency and latency for the rapidly growing Edge AI sector.
Empowering Real-Time AI at the Edge
Traditional AI inference at the network edge has long been bottlenecked by the stringent power constraints of localized devices. EdgeAI’s neuromorphic architecture solves this by mimicking the human brain’s neural structure, performing computation and storage simultaneously while processing data only when an event changes.
Andes Technology’s RISC-V CPU solutions deliver high-performance, low-power intelligence to AI MCU applications by combining control tasks and AI acceleration into a single efficient architecture with the RISC-V Vector (RVV) and Packed SIMD (RVP) support. Furthermore, the Andes Custom Extension (ACE) empowers developers to design custom hardware instructions that eliminate specific algorithmic bottlenecks. This hardware flexibility is backed by the “AndesAire” NN Software libraries, and FreeRTOS, Zephyr OS support.
EdgeAI is currently developing its 4TOPS-class Edge AI accelerator SoC “EGAI_001S” and the event-driven ultra-low-power neuromorphic NPU “EGAI_01N.” Through this license agreement, the company aims to further optimize the performance and power efficiency of its AI semiconductor platform. The joint platform will directly address the critical demands of four major ecosystems requiring real-time, zero-latency intelligence: including industrial IoT, smart infrastructure, intelligent CCTV systems, and future mobility solutions.
Executive Commentary
Kim Jung-wook, Co-CEO of EdgeAI, stated, “We believe the on-device AI and Edge AI markets will become core growth sectors of the future AI industry. This collaboration with Andes will serve as an important milestone for EdgeAI in securing globally competitive RISC-V-based AI semiconductor technologies.
He added, “As demand for low-power and high-efficiency AI semiconductors continues to grow rapidly, EdgeAI plans to accelerate its entry into the global Edge AI market through advancements in its proprietary SoC and neuromorphic NPU technologies.”
“With billions of Andes-powered AI SoCs deployed worldwide, we have built a formidable track record across the global AI landscape. We are excited to channel this extensive expertise into our collaboration with EdgeAI. By providing a highly efficient and customizable computational foundation, we will help EdgeAI accelerate the commercialization of South Korea’s first industrial neuromorphic chip, delivering the strict sub-watt power efficiency and immediate responsiveness next-generation edge applications demand.” said Dr. Charlie Su, President and CTO of Andes Technology.