{"id":12544,"date":"2024-06-07T09:48:57","date_gmt":"2024-06-07T01:48:57","guid":{"rendered":"https:\/\/www.1ai.net\/?p=12544"},"modified":"2024-06-07T09:48:57","modified_gmt":"2024-06-07T01:48:57","slug":"%e6%8c%91%e6%88%98%e8%8b%b1%e4%bc%9f%e8%be%be-amd%ef%bc%8c%e8%83%8c%e9%9d%a0%e9%ab%98%e9%80%9a%e5%92%8c%e5%af%8c%e5%a3%ab%e5%ba%b7%e7%9a%84%e5%88%9d%e5%88%9b%e5%85%ac%e5%8f%b8%e8%80%90%e8%83%bd","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/12544.html","title":{"rendered":"Challenging Nvidia and AMD, Kneron, a startup backed by Qualcomm and Foxconn, launches new NPU for notebooks"},"content":{"rendered":"

AIchip<\/a>Startups<\/a>Nerve<\/a>\uff08Kneron<\/a>) launched its next-generation AI products at the 2024 Taipei International Computer Show on Wednesday.\u00a0KNEO 330 server and PC device equipped with the third-generation NPU chip KL830<\/span>.<\/p>\n

\"ae8461a6-7a2a-4cf1-b31a-9d3726f88c08\"<\/p>\n

It is reported that the company was founded in 2015 by CEO Liu Juncheng and Zhang Maozhong. It is a company that provides edge computing artificial intelligence (edge AI) technology. Its investors mainly include Li Ka-shing'sHorizons Ventures, Qualcomm, Foxconn Group, Alibaba Entrepreneurs Fund, China Development Capital<\/span>And other well-known groups.<\/p>\n

At the press conference, Kneron CEO Liu Juncheng said that companies that currently use generative AI functions and applications rely on cloud giants such as Microsoft and Amazon and purchase a large number of chips from companies such as Nvidia and AMD. For reasons of data privacy and security, companies may be reluctant to hand over their data to external companies and use their AI tools for model training.Therefore, we can use Kneron's technology to build internal servers for enterprises to build enterprise-specific AI applications.<\/span>He hopes that Kneron can profit from the enterprise's AI business and become an alternative to giants such as Nvidia and AMD.<\/p>\n

A total of two products were released at this press conference - the KNEO 330 server and a PC device equipped with the third-generation NPU chip KL830. The official also plans to launch the self-developed fourth-generation NPU chip KL1140 in 2025.<\/p>\n

KNEO 330 server has 48TOPS AI computing power and can support up to 8 concurrent connections<\/span>, supports LLM and Stable Diffusion. Under lower hardware conditions, its RAG accuracy is comparable to cloud solutions, reducing the overall AI cost of small enterprises by 30%~40%.<\/p>\n

The PC device equipped with the third-generation NPU chip KL830 did not disclose specific details. A HP device equipped with the third-generation NPU chip KL830 was used for demonstration at the press conference. Kneron said that it is currently working closely with several of the largest PC OEM manufacturers, who will become the first customers of its latest chips.<\/p>\n

The KL830 chip is a neural processing unit (NPU) that works with the central processing unit (CPU) and graphics processing unit (GPU) to run artificial intelligence (AI) tasks.It can provide up to 10eTOPS@8bit comprehensive computing power, with a peak power consumption of only 2W<\/span>, supporting many large language models with relatively low parameters.<\/p>\n

Kneron said the KL830 can also be used via a USB AI accelerator stick, allowing any device, whether it is a broadband router, IoT camera or traditional computer, to become an edge AI device.<\/p>","protected":false},"excerpt":{"rendered":"

Artificial intelligence chip startup Kneron unveiled its next-generation AI products -- the KNEO 330 server and PC devices equipped with the third-generation NPU chip KL830 -- at Computex Taipei 2024 on Wednesday. Founded in 2015 by CEOs Junsheng Liu and Maozhong Zhang, the company is said to be a provider of edge AI technology, with investors including Li Ka-shing's Harbor Investment, Qualcomm, Hon Hai Group, Alibaba Venture Fund, China Development Capital and other well-known groups. At the launch event, Endurance CEO Liu Juncheng said that companies currently using generative AI features and applications rely on giants such as Microsoft and Amazon for their cloud business, and get their cloud services from NVIDIA<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[2959,309,2960,238],"collection":[],"class_list":["post-12544","post","type-post","status-publish","format-standard","hentry","category-news","tag-kneron","tag-309","tag-2960","tag-238"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/12544","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=12544"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/12544\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=12544"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=12544"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=12544"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=12544"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}