"Starting from this chip" Chip Technology released its AI PC strategy and first chip

"Starting from this chip" Cixin Technology released its AI PC strategy and first chip Laying out the end-side AI ecosystem Cixin Technology released heterogeneous AI PC chips

On July 30, with the theme of "Starting from this chip", the AI ​​PC strategy and first chip launch conference of Cixin Technology was held in Shanghai. Facing the coming edge-side generative AI era and the third PC industry revolution, Cixin Technology has established a development strategy of "one chip for multiple uses", focusing on the AI ​​PC field first, creating a new generation of AI PC computing power base, and launching the first heterogeneous high-efficiency SoC Cixin P1.

"CiXin Technology is committed to promoting the development of the edge AI ecosystem with high-efficiency computing solutions, and making life, learning, and work smarter, more efficient, and more humane through the application of AI technology." Sun Wenjian, founder and CEO of CiXin Technology, said at the press conference that he hopes to create a better future with AI.

Sun Wenjian, founder and CEO of Cixin Technology

One chip for multiple uses to build a new generation of AI PC computing power base

With the release of ChatGPT, artificial intelligence has entered the era of deep learning 2.0, and generative AI has become the main theme of development. With the in-depth deployment and application of generative AI big models in various industries, users' demands for energy efficiency, data privacy, and personalized customization have become increasingly strong. The hybrid deployment of cloud-based public big models and end-side private big models has become an industry consensus. As the best carrier of end-side generative AI, the PC industry has put forward new requirements for the AI ​​computing power of hardware, especially SoC as the core computing unit.

Based on this, Cixin Technology has determined the development strategy of "one chip for multiple uses", targeting both the global and local markets, building an end-side AI ecosystem, focusing on the AI ​​PC field first, and creating a new generation of AI PC computing power base.

Sun Wenjian said: "The rich functions of the chip of Cixin Technology greatly meet the needs of customers in multiple scenarios; on the other hand, through the implementation of multiple scenarios, the sales volume of products increases, the product R&D costs are diluted, and customers are provided with a cost-effective product experience."

Regarding the new generation of AI PC computing power base, Sun Wenjian introduced: "The AI ​​PC computing power base of Cixin Technology is heterogeneous, energy-efficient and secure." Its heterogeneous integration of CPU, GPU, and NPU achieves efficient operation based on data link bandwidth optimization. In terms of energy efficiency, it adopts multi-core heterogeneous and dedicated NPU design and low-power memory technology, and achieves high energy efficiency through software and hardware collaborative optimization. In terms of security, Cixin P1 has the latest security features such as PACBTI, MTE, and secure EL2 in the Arm®v9 architecture, supports general commercial and national secret algorithms, and meets the requirements of TPM (Trusted Platform Module) and TCM (Trusted Cryptographic Module), providing system-level security and privacy protection.

"Cixin Technology's new generation of AI PC computing power base is aimed at generative AI and supports hybrid artificial intelligence deployment." Sun Wenjian further pointed out that by building a rich open ecosystem of software and hardware, it empowers developers and continues to explore the application scenarios and advantages of edge AI PCs.

The first heterogeneous AI PC chip is released

At the press conference, based on the AI ​​PC strategy, Cixin Technology released its first product, the Cixin P1, a heterogeneous, energy-efficient SoC designed specifically for AI PCs. Sun Wenjian said: "After rigorous testing, Cixin P1 has fully met the requirements for mass production and will officially enter the productization stage."

ChipCore launches its first AI PC chip

This chip P1 uses advanced 6nm manufacturing process, providing rich AI heterogeneous computing resources, a full range of security engines, diverse peripheral interfaces, and multi-operating system support. The powerful multimedia engine supports 4K120 frame display, 8K60 frame video decoding, and 8K30 frame video encoding; the high-performance memory access subsystem is equipped with 128-bit LPDDR5 low-power memory, with a capacity of up to 64GB, a data transmission rate of up to 6400Mbps, and a bandwidth of up to 100GB/s. At the same time, it has efficient power management, providing precise dynamic frequency and voltage regulation, multiple power domains and dynamic power gating, and standard PC power working mode.

The core CPU part is designed with Arm big.LITTLE™ technology, with 8 performance cores and 4 energy-efficiency cores, a main frequency of up to 3.2GHz and a multi-level cache design optimized for PC scenarios; at the same time, it integrates 2 SVE2 vector acceleration units to achieve machine learning instruction enhancement.

The integrated GPU provides a 10-core GPU processor to meet the needs of extreme desktop rendering and general AI computing. The new generation of hardware ray tracing provides a console-level gaming experience; the new geometry processing flow (deferred vertex shading DVS) saves more than 40% of power consumption, and flexible variable speed shading (VRS) improves performance by more than 50%. At the same time, the desktop GPU software stack for multiple scenarios meets the needs of industry applications.

The powerful heterogeneous AI engine provides 45TOPS of terminal AI heterogeneous computing power, supports the deployment of large models on the terminal with less than 10 billion parameters, and can run LLM at more than 30 tokens/s, providing terminal AI support for multiple scenarios such as computer vision, natural language processing, and generative AI.

James McNiven, vice president of product management, Arm Endpoint Business Unit, said: "AI is changing the way consumers use PCs. The P1 chip, developed on Arm technology, makes the advantages of the PC ecosystem based on the Arm platform benefit a wider audience, bringing a high-performance, energy-efficient and AI-optimized platform to redefine the user experience for various workloads and use cases."

Xu Yatao, Executive Vice President of Sales and Business of Arm Technology said: "ChipCore Technology is an important partner of Arm Technology. In recent years, the two parties have jointly promoted the iteration and upgrade of Arm CPU technology and led the expansion of the advanced Armv9 architecture in the PC application ecosystem. In the future, we will continue to maintain close cooperation in multiple key areas such as product research and development and technology ecology, provide the industry with more reliable and efficient computing solutions, and promote the widespread application of terminal-side AI in various fields."

Xu Yatao, Executive Vice President of Sales and Business of ARM Technology

Complete solution supports accelerated AI PC development process

"As a newcomer in the PC chip field, Cixin Technology has the opportunity to create AI PC platform solutions from a new perspective," said Chu Ranzhou, co-founder and vice president of system engineering of Cixin Technology. Cixin Technology's "three-in-one" strategy for platform solutions was born.

Chu Ranzhou, Co-founder and Vice President of System Engineering of Cixin Technology

The "three integrations" strategy is to integrate the advantages of the two major architectures of X86 and Arm, integrate into the circle of friends in the PC industry, and integrate into the world of AI. Based on the "three integrations" strategy, Cixin Technology launched the AI ​​PC platform solution of the P1 chip, which has the characteristics of scalable heterogeneous computing, support for multi-modal human-computer interaction, high-bandwidth storage, and platform-level security shield.

In addition, the Cixin Technology AI PC platform solution supports a variety of interfaces including Video-In/out, GMAC/Ethernet, HDA/I2S, etc., providing a foundation for universal and inclusive AI; at the same time, based on the "one chip for multiple uses" strategy, Cixin P1 will launch a variety of specifications to support the implementation of various product forms of AI terminals. As one of the factors of PC product cost, the type of PCB is related to the complexity and yield of SMT process. The Cixin P1 platform solution can support all types of 8-12 layers, through-hole and high-density PCBs, eliminating the trouble of customers having to select PCB suppliers and SMT foundries when making products.

Chu Ranzhou said: "We are actively integrating into the PC industry chain and can support PC manufacturers to seamlessly switch from X86 CPU to this P1 chip. We have also started cooperation with mainstream ODM manufacturers and well-known BIOS manufacturers in the industry."

In terms of customer project support, Chu Ranzhou introduced that Cixin Technology has established a customer support system "based on comprehensive linkage among marketing, support, and R&D teams to quickly respond to customer requirements, implement application scenarios, quickly resolve problems, and shorten product development cycles."

Using software to promote hardware and promote innovation in the AI ​​PC industry

"Hardware is like our body, and software gives a chip a soul." When talking about AI PC software solutions, Liu Gang, co-founder and vice president of software engineering of Cixin Technology, said.

Gang Liu, Co-founder and Vice President of Software Engineering of Cixin Technology