Core Insights - The article highlights the launch of the scaleX 30,000-card supercluster by Zhongke Shuguang, marking the establishment of the largest domestic AI computing power pool in China, which supports advanced AI computing scenarios such as trillion-parameter model training and AI for Science [1][5]. Group 1: Technological Breakthrough - The scaleX supercluster was deployed in less than two months after its first appearance at the HAIC conference in December 2025, showcasing China's rapid integration capabilities in large-scale intelligent computing systems [2]. - Zhongke Shuguang overcame several key technical challenges, including high-speed interconnection, tight coupling design of storage and computing, and efficient resource scheduling, through systematic innovation [4]. Group 2: Practical Value - The scaleX supercluster has demonstrated its application value across various fields, including supporting trillion-parameter model training and enhancing inference efficiency for major internet companies [5]. - It has also contributed to significant advancements in scientific research, such as improving protein research efficiency by 3 to 6 orders of magnitude for top domestic research teams [5]. Group 3: Regional Empowerment - The simultaneous deployment of three superclusters is expected to significantly impact regional development, particularly in Zhengzhou, enhancing its capabilities in computing power and digital infrastructure [6]. - The establishment of this core node is anticipated to elevate Zhengzhou's development level and strengthen its role in computing power scheduling and multi-domain applications [6].

Core Insights - The National Supercomputing Internet Core Node has launched in Zhengzhou, marking the first operational deployment of a 30,000-card domestic AI computing pool in China [1] - The scaleX 10,000-card super cluster is a significant advancement in AI computing, achieving a 20-fold increase in cabinet-level computing density and a 30% to 40% performance improvement for MoE trillion-parameter model training and inference [1][2] - The scaleX640 super node, which supports multi-brand acceleration cards and is compatible with mainstream computing ecosystems, aims to lower the R&D threshold for AI clusters and avoid redundant investments [2] Technological Advancements - The scaleX640 super node features an "open architecture" that enhances communication speed and reduces latency, enabling the formation of thousand-card computing units [1][2] - The supercomputing pool supports various large-scale AI computing scenarios, including trillion-parameter model training, high-throughput inference, and AI for Science applications, significantly improving research efficiency in fields like protein studies [2] Future Developments - The company plans to establish an Advanced Computing Technology Research Institute in Zhengzhou, with a 160,000 square meter R&D space and an initial goal of gathering 1,000 researchers focused on AI and high-performance computing applications [3]

Group 1 - The core viewpoint of the article is the launch of China's first one-stop development platform for scientific large models, named OneScience, by Zhongke Shuguang at the 2025 China Computer Conference [2] Group 2 - The platform aims to facilitate the development and deployment of large-scale scientific models, indicating a significant advancement in China's capabilities in artificial intelligence and computational science [2] - This initiative reflects the growing trend in the industry towards integrating AI technologies into various scientific fields, enhancing research efficiency and innovation [2] - The launch of OneScience positions Zhongke Shuguang as a key player in the competitive landscape of AI development platforms in China [2]