Kneron Debuts Edge AI Chip, Bringing AI to Devices Everywhere
Amidst the fierce competition in the world of artificial intelligence dominated by global tech companies, an AI startup in Taiwan has emerged as the pioneer in the field of edge AI. Kneron was founded by a young Taiwan native Albert Liu, and received funding from renowned investors including Himax, Qualcomm, CDIB, Alibaba Entrepreneurs Fund and Horizons Ventures, the private technology investment arm of Mr Li Ka-Shing. Today, Kneron is unveiling its ground-breaking 3D artificial intelligence solution and edge AI chip.
Named after its release date of May 20th, Kneron’s debut AI Chip “KL520” has the best-in-class power efficiency only consuming a few hundred mega-watts, bringing AI computation power to edge devices for various applications.
Building Global Partnerships to Bring AI to the Masses
Further to the KL520 edge AI chip release, Kneron is also announcing partnerships with industry leaders to bring their chip into our everyday lives. Partnerships include fabless IC maker Etron Technology Inc., US-listed company Himax Technologies Inc., enterprise PC OEM AAEON Technology Inc., 3C solution provider Alltek Technology Corp, renowned ODM players Pegatron, as well as 3D sensing provider Orbbec, and IC design house Datang Semiconductor.
Founded in 2015, Kneron is headquartered in San Diego. Founder Albert Liu graduated from National Cheng Kung University, received his PhD from UCLA, and subsequently worked at Qualcomm, Samsung Electronics and MStar, developing advanced AI systems.
Proprietary Reconfigurable Architecture to Bring Efficient AI Solutions On-Device
The KL520 edge AI chip is a culmination of Kneron’s core technologies, combining proprietary software and hardware designs to create a highly efficient and ultra low-power Neural Processing Unit (NPU). Running AI computations on the end device will help generate real-time insights without relying on the cloud.
In particular, KL520 is equipped with a unique reconfigurable architecture which allows the chip to run a number of different convolutional neural networks (CNN) based on different applications, regardless of its kernel size, architecture requirements or input size. In turn, the chip is able to achieve a very high MAC (media access control) efficiency, a key metric for efficient computation. This software innovation allows the chip to dynamically adjust its function based on applications needs, in order to achieve maximum efficiency. The KL520 is also equipped with a proprietary compression technology that is able to compress a large AI neural network into a very small footprint to process on the edge device, without sacrificing large amounts of power.
Kneron’s software and hardware innovations are truly a breakthrough in the development of edge AI. By bringing elements of reconfigurability and compression onto the chip, the KL520 is bringing AI computation to various end-devices that operate under low-powered environments.
Features and targeted applications for the KL520:
- Low-powered with a small physical footprint
- The KL250 can run alongside a main chip as a co-processor; will not need a replacement chip
- For smart door lock applications, KL520 includes two ARM Cortex M4 CPU, which can serve as the main processor.
- Balances the need of performance, power and cost to bring the best solution for edge applications
- Applicable to various 3D sensor technologies such as structured light, dual-camera and ToF, and Kneron’s exclusive 3D sensing technology
- Well-suited for applications including smart locks, security cameras, drones, smart home appliances and robotics
During the product launch event, Kneron’s Founder and CEO, Albert Liu said: “We are determined to deliver the best-in-class edge AI solution to the market. The results that we have achieved over the past four years are very promising, and we are proud to receive recognitions from industry partners for our innovations.”