. Billions of yuan! The last AI chip financing this year has arrived - Zhuoliou industry co.,ltd
+86 13640908945+86 13640908945 sales@chipscomponents.comsales@chipscomponents.com

Welcome to Zhuoliou industry co.,ltd

Billions of yuan! The last AI chip financing this year has arrived

3/9/2023 11:13:53 AM

Shenzhen reconfigurable data flow AI chip startup Kunyun Technology has announced the completion of a Series C financing of hundreds of millions of yuan, led by Pro Capital, and co invested by Dinghui Baifu, Unicom's Lianchuang Fund, Zhang Keyaokun Fund, and Zhonglou Financial Holding Group.


Kunyun Technology focuses on underlying technological innovation and develops high-performance, low-cost AI chips with high computing power based on the reconfigurable data flow technology route. Reconfigurable data flow architecture is a computational architecture approach that defines computation order based on reconfigurable spatial connectivity relationships.


The new financing will mainly be used to support the research and development and scale implementation of Kunyun Technology's next-generation reconfigurable data flow CAISA AI chips, and to build an industrialization ecosystem for CAISA chips in various vertical industries.


1\ The founder graduated from Imperial College of Technology and studied under Academician Lu Yongqing of the Royal Academy of Engineering in the UK

Shenzhen Kunyun Technology was established in 2017, and most of the early team members were from Imperial College London.

Dr. Niu Xinyu, the founder and CEO of the company, graduated from the Department of Electronic Engineering at Fudan University with a bachelor's degree. He then went to Imperial College London for further studies under the guidance of Wayne Luk, a professor at Imperial College London, a member of the Royal Academy of Sciences, an IEEE member, and a BCS member.


2\ Self developed reconfigurable data flow architecture AI chip, with products serving over 700 information technology enterprises
Unlike the instruction set architecture, which is based on the temporal execution order of instructions, the reconfigurable data flow architecture does not include instructions. By modifying the connection relationship of computing units in space to control the calculation order, the redundancy brought by instruction read and write operations can be optimized, achieving higher chip utilization and meeting the computing power needs of customers without pursuing high process technology.

In 2020, Kunyun Technology launched the world's first reconfigurable data stream architecture AI chip, CAISA. Compared with similar international products, it achieved an order of magnitude increase in chip utilization indicators, with a maximum chip utilization rate of 95.4% and a latency of less than 3ms.

With a new technology roadmap chip based on reconfigurable data flow architecture, Kunyun Technology has won multiple top awards in the field of artificial intelligence, including the first prize of the Wu Wenjun Artificial Intelligence Science and Technology Award, the highest award of the World Artificial Intelligence Conference, the SAIL Award for Outstanding Artificial Intelligence Leader, and the Ministry of Industry and Information Technology's "China Chip" Excellent Technology Innovation Product Award.
Up to now, a series of computing power products based on CAISA chips have covered a full range of computing power products, including acceleration cards, edge small stations, intelligent computing hosts, intelligent computing servers, and more.
These products have served more than 700 information technology enterprises, including China Mobile, China Unicom, China Telecom, China Tower, Inspur Group, Central Control Technology, and China Electronics Technology. The delivered computing power products support customers in stable delivery of over 20 industries and thousands of digital projects, such as petrochemical, mining, power, and urban lifelines.

3\ Improving the cost-effectiveness and ease of use of computing power, the underlying innovative architecture is favored by investors
Niu Xinyu, founder and CEO of Kunyun Technology, said that Kunyun's mission is to provide the next generation computing platform and accelerate the implementation of artificial intelligence.
Among them, "providing the next generation computing platform" means constantly iterating the CAISA chip product line, improving computing power, cost-effectiveness, and ease of use through a new reconfigurable data flow technology route; Accelerating the implementation of artificial intelligence means continuously building an ecosystem that creates value for customers and benefits ecological partners.
Niu Xinyu said that after multiple generations of product iterations, the CAISA series chips have gained increasing recognition and support from customers. Market feedback has proven that this is a feasible path that can provide higher performance, lower costs, and create value for customers.
Pro Capital, Dinghui Baifu, Unicom's Lianchuang Fund, Zhang Keyaokun Fund, and Zhonglou Financial Holding Group, which participated in the C-round financing of Kunyun Technology, have all expressed their recognition of Kunyun Technology's accumulation of reconfigurable data flow architecture technology from underlying architecture innovation.
In the view of Xu Chenhao, the executive partner of Pro Capital, a reconfigurable data flow architecture is currently one of the best solutions that can significantly improve the cost-effectiveness of computing power and solve the bottleneck problem of advanced technology. As a leading enterprise in this field, Kunyun Technology's excellent product performance and rapid commercialization results are the best annotations to this viewpoint.
Ying Wei, Managing Partner of Dinghui Baifu, referred to Kunyun Technology as a "scarce target for reconfigurable data flow architecture AI chips in China" and expressed optimism that Kunyun can empower multiple application scenarios with high cost performance solutions.
As the national team of big data computing power platforms, China Unicom attaches great importance to the ecological layout of the computing power track. Xu Baiming, General Manager of Unicom's Lianchuang Fund, commented that Kunyun Technology's reconfigurable data flow architecture technology has taken a new path in the vertical track, providing new choices for the underlying computing power support of the current digital economy development.
Gu Minqiang, Investment Director of Shanghai Zhangke Yaokun Venture Capital Partnership (Zhangke Yaokun Fund), stated that Kunyun Technology's reconfigurable data flow architecture can reduce the dependence of chips on advanced processes under the same computing power requirements, and has a strong advantage in product cost-effectiveness. At the same time, their products have been applied in multiple scenarios and have been preliminarily validated in the market, which is the main reason why they chose Kunyun.


Yan Jiandong, Chairman of Zhonglou Financial Holding Group, mentioned that Kunyun Technology has created a comprehensive AI implementation plan in closed application scenarios such as coal mines, providing strong support for accelerating digital transformation of the leading industries in Zhonglou District, improving production efficiency, and optimizing industrial structure.

Blog Category

News

Latest Products

IP5002CX8/P135

IP5002CX8/P135

NXP USA Inc.
ADAU7002ACBZ-RL

ADAU7002ACBZ-RL

Analog Devices Inc.
PGA2320IDW

PGA2320IDW

Texas Instruments
SRC4184IPAG

SRC4184IPAG

Texas Instruments
MUSES72320V-TE2

MUSES72320V-TE2

Nisshinbo Micro Devices Inc.
PCM2706CPJT

PCM2706CPJT

Texas Instruments
ZL38040LDG1

ZL38040LDG1

Microchip Technology
PGA2310UA/1K

PGA2310UA/1K

Texas Instruments
Help you to save your cost and time.
Help you to save your cost and time.
Reliable package for your goods.
Reliable package for your goods.
Fast Reliable Delivery to save time.
Fast Reliable Delivery to save time.
Quality premium after-sale service.
Quality premium after-sale service.
Copyright © 2023 Zhuoliou industry co.,ltd
HOME

HOME

PRODUCT

PRODUCT

PHONE

PHONE

USER

USER