【清华大学芯片研究实现重大突破，光计算架构开启高性能算力新篇章】
清华大学在芯片研究领域取得了重要进展，为高性能计算领域带来创新灵感。据11日的最新消息，该校电子工程系副教授方璐的课题组与自动化系戴琼海院士的课题组，针对大规模光电智能计算的挑战，成功研发出分布式广度光计算架构，并以此为基础，创新性地设计出名为“太极”（Taichi）的大规模干涉-衍射异构集成芯片。
这一突破性成果摒弃了传统的电子深度计算模式，转而采用光子计算，实现了160 TOPS/W的通用智能计算效能，这在当前大模型通用人工智能快速发展的背景下，无疑为高性能算力的提升开辟了新的道路。这一创新成果的出现，预示着光子计算可能成为未来提升计算效率的关键技术。
“太极”芯片的科研成果已正式发表在国际知名期刊《科学》的最新一期，彰显了中国在高科技领域的研究实力和创新能力。这一突破不仅在学术界引起广泛关注，也为工业界提供了新的技术参考，有望推动全球芯片技术和人工智能应用的进一步发展。
英语如下：
**News Title:** “Tsinghua University Breaks Ground in Chip Research with Optical Computing Chip ‘Taichi’ Featured in ‘Science’ Journal”
**Keywords:** Tsinghua University, chip breakthrough, optical computing
**News Content:**
**Tsinghua University Makes Major Chip Research Breakthrough, Paving the Way for High-Performance Computing with Optical Computing**
Tsinghua University has achieved a significant advancement in chip research, inspiring innovation in the high-performance computing domain. According to the latest news on the 11th, the research group led by Associate Professor Fang Lu from the Department of Electronic Engineering, in collaboration with Academician Dai Qionghai’s group from the Department of Automation, successfully developed a distributed broadband optical computing architecture. Based on this, they creatively designed a large-scale interference-diffraction heterogeneous integration chip named “Taichi.”
This breakthrough departure from conventional electronic deep computing methods adopts photonic computing, realizing a general-purpose intelligent computing efficiency of 160 TOPS/W. In the context of the rapid development of large-scale AI models, this milestone opens up a new path for enhancing high-performance computing capabilities. The emergence of this innovative result suggests that photonic computing could be a crucial technology for boosting computational efficiency in the future.
The research findings on the “Taichi” chip have been formally published in the latest issue of the renowned international journal ‘Science,’ demonstrating China’s research prowess and innovation capacity in high-tech fields. This breakthrough has attracted widespread attention in academia and is poised to provide new technological references for the industry, potentially fueling further progress in global chip technology and AI applications.
【来源】https://www.cls.cn/detail/1644516
Views: 1