Chinese search giant Baidu has announced a new AI-focused processor dubbed Kunlun, which the company claims is capable of handling both datacentre and edge workloads. Kunlun will include the 818-300 training chip, and the 818-100 for inference.
The company said it started developing a field-programmable gate array AI accelerator in 2011, and that Kunlun is almost 30 times faster. The chips are made with Samsung's 14nm process, have 512GBps memory bandwidth, and are capable of 260 tera operations per second at 100 watts.
"[Kunlun] leverages Baidu's AI ecosystem, which includes AI scenarios like search ranking and deep learning frameworks like PaddlePaddle," the company said. "Baidu's years of experience in optimising the performance of these AI services and frameworks afforded the company the expertise required to build a world-class AI chip."
"In addition to supporting the common open-source deep learning algorithms, Kunlun chip can also support a wide variety of AI applications, including voice recognition, search ranking, natural language processing, autonomous driving, and large-scale recommendations," Baidu said.
Given the recent threat to the ongoing existence of ZTE from a seven-year US-imposed ban on the importation of American components, which was eventually rescinded by US President Donald Trump[1], China is looking to ramp up its domestic processor production.
Speaking at the end of May, founder and CEO of Tencent, Pony Ma, said the ZTE furore showed China lacking key technologies[2] such as processors and operating systems.
Ma said Tencent was looking to support the development of China's chip industry.
The Chinese government is looking to its Made in China 2025 plan to have domestic chips make up 40 percent of its needs.
Earlier, Intel had announced that Baidu would