全球领先，国产万卡超集群首次真机亮相

Core Viewpoint - The launch of the scaleX WanKa supercluster by Zhongke Shuguang at the HAIC2025 marks a significant advancement in large-scale intelligent computing systems, showcasing China's capabilities in AI infrastructure [1][3]. Group 1: Product Features - The scaleX WanKa supercluster is designed for trillion-parameter models and complex tasks, featuring innovations in performance, efficiency, reliability, and scalability [3]. - It includes the world's first single-cabinet 640-card supernode, achieving over 5 EFlops of total computing power and a PUE value as low as 1.04, enhancing computing density by 20 times [3][4]. - The system utilizes a self-developed RDMA high-speed network, achieving 400 Gb/s bandwidth and under 1 microsecond communication latency, which improves performance by 2.33 times compared to traditional networks [4]. - The architecture incorporates deep optimization for storage, computing, and transmission, enhancing AI inference speed and accuracy while increasing resource utilization by 55% [4]. - It features a digital twin for intelligent scheduling and management, ensuring 99.99% availability and supporting large-scale job scheduling [4]. Group 2: Market Impact and Collaboration - The scaleX WanKa supercluster supports multiple brands of acceleration cards and mainstream computing ecosystems, optimizing over 400 mainstream models [6]. - It is part of the "AI Computing Open Architecture," developed in collaboration with over 20 AI industry chain companies, aimed at lowering the R&D threshold for intelligent computing clusters [6]. - The initiative promotes a shift from isolated technological breakthroughs to collaborative industry advancements, translating open concepts into practical computing power [6].