Kneron launches the latest AI chip KL730, driving widespread adoption of lightweight GPT solutions
Views : 515
Update time : 2023-08-21 12:10:39
On August 17, Kneron recently released the KL730 chip. The KL730 integrates a vehicle-grade NPU and an image signal processor (ISP), and empowers safe and low-energy AI capabilities to edge servers, smart homes, and car-assisted driving systems. In various application scenarios.
As the Kneron chip, the KL730 has been designed to realize AI functions from the start, and has broken through a number of technological innovations in energy saving and security. The chip has a multi-channel interface, which can seamlessly access various digital signals such as image, video, audio and millimeter wave, and supports the development of artificial intelligence applications in various industries. The chip also solves one of the current widespread bottlenecks of artificial intelligence: the high system cost caused by widespread low-energy hardware. In the research and development of this chip, KL730 has made great progress in energy efficiency. The energy efficiency ratio is 3 to 4 times higher than that of previous chips, and it is 150%25 to 200%25 higher than that of equivalent products in major industries. Juncheng Liu, founder and CEO of Kneron, said: "Running AI functions requires dedicated AI chips, and the architecture is completely different from what we have known before. Simply reusing adjacent technologies (such as GPU chips dedicated to graphics processing) will not do the job. KL730 will be an AI innovator at the edge. With its unprecedented efficiency and support for frameworks such as Transformer, we are providing thousands of industries with powerful AI capabilities to fully realize the potential of artificial intelligence with extreme security and data privacy." In 2021, Kneron released KL530, the first edge AI chip to support Transformer neural network architecture. The Transfomer neural network architecture is the foundation of all GPT (Generative Pre-trained Transformer) models. The KL730 chip further enriches the product family, providing 0.35 - 4 tera of effective computing power per second, extending the ability to support state-of-the-art lightweight GPT large language models such as nanoGPT. The KL730 is uniquely positioned to transform AIot security, enabling users to run GPT models partially or completely offline at the end point. The chip, in conjunction with Kneron's private secure edge AI network Kneo, allows AI to run on users' edge devices, giving them better data privacy. These applications are industry-wide: from Enterprise Services solutions to intelligent vehicles to AI-powered medical devices, the added security allows for greater collaboration between devices while keeping data secure. Since its founding in 2015, Kneron has received numerous industry accolades and awards for its reconfigurable NPU architecture. For example, the IEEE Darlington Award. Endpoint products using Nevermind chips have entered many industries, from AIoT to intelligent driving and edge servers, including Toyota, Quanta Electronics, Chunghwa Telecom, Panasonic, Hanwha and many others.
The LDK220 series of low-dropout linear regulators (LDOs) is a high-performance device designed specifically for low-power consumption and high-precision voltage regulation, widely used in scenarios such as consumer electronics, industrial control, and po
Xilinx Spartan®-7 FPGA Family stands as a defining solution in the mid-range FPGA landscape, blending high performance, energy efficiency, and cost-effectiveness to redefine versatility for industrial, IoT, and consumer electronics applications. Built on
The Altera FLEX series was more than a lineup of FPGAs—it was a blueprint for how programmable logic devices could evolve to meet diverse market needs. The FLEX 8000 laid the architectural groundwork, the FLEX 10K redefined functionality with embedded mem
Among TI standout offerings, the LM4765 and LM4766 are dual-channel amplifiers designed to cater to diverse audio needs—from compact setups to high-fidelity systems. While sharing the same product lineage, these chips differ significantly in power output,
