再融资新政后沪市首单 中科曙光拟发不超80亿可转债

Core Viewpoint - The issuance of convertible bonds by Zhongke Shuguang, amounting to up to 8 billion yuan, marks the first proposal following the new refinancing policy in the Shanghai market, aimed at supporting high-quality listed companies in innovation and development [1][2]. Group 1: Financing and Investment - Zhongke Shuguang plans to use the funds raised from the convertible bonds entirely for three core AI computing projects: advanced computing cluster systems, AI training and inference integrated machines, and domestic storage systems [1][3]. - The company has successfully developed the scaleX640 super node, achieving high-speed interconnection with 640 cards in a single cabinet, and has overcome key technologies related to high-speed interconnection and tight coupling of storage and computing [3][4]. - The next-generation high-performance AI training and inference integrated machine aims to address the challenges of deploying AI in various industries, providing a fully optimized solution that reduces deployment barriers and long-term operational costs [3][4]. Group 2: Market Context and Demand - The global AI computing demand is surging, with the AI server market expected to grow from approximately $125.1 billion in 2024 to $158.7 billion in 2025, while China's market is projected to reach $25.9 billion, reflecting a year-on-year growth of 36.32% [9]. - The national strategy emphasizes the need for domestic innovation in storage systems, which are critical for AI's large-scale development, focusing on performance, latency, concurrency, and reliability [4][9]. Group 3: Financial Performance and Projections - For the first three quarters of 2025, Zhongke Shuguang reported revenues of 8.82 billion yuan and a net profit of 966 million yuan, with a projected net profit of approximately 2.465 billion yuan for the full year, indicating a year-on-year growth of 28.99% [6][7]. - The company’s asset-liability ratio is expected to decrease post-issuance of the convertible bonds, leading to an improved financial structure [5][6]. - Research and development expenses are projected to reach 1.292 billion yuan in 2024, accounting for 9.83% of revenue, reflecting the company's commitment to innovation [7]. Group 4: Technological Advancements - The advanced storage system developed by Zhongke Shuguang has achieved industry-leading performance metrics, including millions of IOPS and sub-millisecond latency, which are essential for handling large-scale AI data processing [4][10]. - The company’s AI computing cluster has been successfully deployed, supporting over 400 mainstream models and achieving significant breakthroughs in AI for scientific research [10].