Reading Good Books Together
Recently, two domestic GPUs have made new progress, namely the new generation “Fuxi” architecture chip from Xiangdi and the 7G100 chip from Lishuan Technology.
The New Generation “Fuxi” Architecture Chip from Xiangdi Has Completed Tape-Out Verification
On September 3, Anfu Technology stated on its interactive platform that the new generation “Fuxi” architecture chip developed by Xiangdi has completed tape-out verification, demonstrating excellent performance in graphics rendering and parallel computing.
It is reported that after receiving investment from Anfu Technology, Xiangdi has accelerated its product iteration, and the new generation Fuxi architecture GPU will adopt a 5nm process, achieving a computing power of 160 TFLOPS (FP32) and integrating 12GB of HBM2 memory.
In mid-April this year, as a shareholder of Xiangdi, Anfu Technology revealed the latest progress of Xiangdi on its interactive platform, stating that the new generation “Fuxi” architecture chip is currently in tape-out, and this architecture will bring two new products, both of which meet the leading domestic performance and technical specifications. The Fuxi A0 focuses on filling the gap in domestic high-end rendering products; the Fuxi B0 is a chip that integrates GPU and NPU, targeting edge model deployment and the rapidly emerging AIPC market, with full support for the edge deployment needs of mainstream models such as LLAMA, ChatGLM-6B, Stable-Diffusion, Sora, and DeepSeek R1 1.5B/7B.
Anfu Technology also revealed that among Xiangdi’s shareholders are investment institutions with state-owned backgrounds, such as Chongqing Liangjiang Industrial Development Group. The listing plan for Xiangdi will be decided based on market conditions, development strategies, and other factors. The company will continue to monitor Xiangdi’s development dynamics and actively support its growth.
Dongxin Co., Ltd.: Lishuan Technology is Sending Samples to Some Customers
On September 3, Dongxin Co., Ltd. stated on its interactive platform that as of now, Lishuan Technology has sent samples to some customers and is continuously optimizing, and will continue to advance product production, sales, and other related work as planned.
According to information, Lishuan Technology was established in 2021 and is a company that develops high-performance GPUs, focusing on self-developed architectures and fully owned intellectual property rights, deeply mastering the architecture, design, software development, and know-how of large GPU chips, with profound experience in processor ecosystem conditions and ecosystem construction.
The 6nm 7G100 chip from Lishuan Technology has undergone nearly four years of research and development, and the chip successfully lit up on May 25, 2025, with the main functional test results meeting expected requirements. The product can cover mainstream graphics rendering and AI acceleration needs in edge, cloud, and edge scenarios, and achieves seamless compatibility with mainstream GPU architectures, with the core goal of breaking through the challenges of achieving a fully controllable domestic mainstream GPU architecture.
Not long ago, Dongxin Co., Ltd. also released an announcement on external investment and related transactions, announcing plans to jointly invest with Hengtong Group, Shanghai Daohua’s funds, and other investors, totaling 500 million yuan to increase capital for Lishuan Technology. Among them, Dongxin Co., Ltd. will subscribe to the newly registered capital with its own funds of about 211 million yuan, and after the transaction is completed, its shareholding ratio will be slightly adjusted from 37.14% to 35.87%, still making it an important shareholder of Shanghai Lishuan. This move marks Dongxin Co., Ltd.’s further deepening of investment in the GPU field within its integrated strategy of “storage, computing, and networking.”
Source: Daily IC