Group 1 - The CPO concept stocks in the A-share market experienced a widespread decline, with notable drops in several companies [1] - Yongding Co., Ltd. saw a significant decrease of over 8%, while Fenghuo Communication and Zhongjing Electronics fell by more than 4% [1] - Other companies such as Sry New Materials, Benchuan Intelligent, Changxin Bochuang, Jufei Optoelectronics, and Tongfu Microelectronics also reported declines exceeding 3% [1] Group 2 - Yongding Co., Ltd. reported a decline of 8.63% with a total market value of 35.9 billion [2] - Fenghuo Communication decreased by 4.80%, with a market capitalization of 51.2 billion, and has a year-to-date increase of 17.52% [2] - Zhongjing Electronics fell by 4.63%, with a market value of 6.935 billion, and a year-to-date decline of 7.06% [2] - Sry New Materials decreased by 3.91%, with a market capitalization of 30.5 billion, and a year-to-date increase of 1.55% [2] - Benchuan Intelligent saw a decline of 3.79%, with a market value of 4.263 billion, and a year-to-date decrease of 0.81% [2] - Changxin Bochuang dropped by 3.49%, with a market capitalization of 43.4 billion, and a year-to-date increase of 4.72% [2] - Jufei Optoelectronics declined by 3.37%, with a market value of 9.748 billion, and a year-to-date increase of 0.58% [2] - Tongfu Microelectronics fell by 3.26%, with a market capitalization of 75.6 billion, and a year-to-date increase of 32.07% [2] - Other companies such as Ziguang Co., Huamao Technology, and Kechuan Technology also reported declines, with varying market values and year-to-date performances [2]

Investment Rating - The industry investment rating is maintained as "Positive" [2] Core Insights - The report emphasizes the transition to the "SuperNode" era in computing infrastructure, driven by the increasing computational demands of AI models, which are evolving from pre-training to multi-faceted scaling [14][17] - The SuperNode architecture relies on enhanced interconnectivity, integrating multiple computing chips into a logical "super GPU/ASIC" to overcome the limitations of traditional single-server setups [39] - The report identifies three main investment themes: "Network End + AIDC + Computing End" and four key sectors: "Optical + Liquid Cooling + Power Supply + Chips" [7] Summary by Sections 1. Transition to SuperNode Era - The evolution of AI models necessitates a shift in computational infrastructure, with demands for higher performance and efficiency [14] - The report highlights the rapid growth in model parameters and training data, indicating a need for advanced computational clusters [14] 2. SuperNode Architecture - SuperNode clusters are defined as large-scale GPU/ASIC systems that expand from thousands to millions of computational units, utilizing both Scale Up and Scale Out strategies [17] - The architecture includes components such as computing nodes, switch trays, and power supply units, which are essential for the operation of SuperNode systems [80] 3. Upgrades in Key Components - The report notes that the increasing penetration of SuperNode servers will drive demand for high-power supplies, liquid cooling systems, and optical communication technologies [5][80] - Specific companies are highlighted as beneficiaries in the sectors of optical modules, liquid cooling, server power supplies, and exchange chips [7] 4. Domestic SuperNode Development - The report discusses the progress of domestic SuperNode solutions, particularly Huawei's Atlas 900 A3 and Atlas 960, which are designed to enhance the performance of domestic AI clusters [6][112] - The performance of domestic chips is shown to improve significantly when deployed in SuperNode configurations, effectively compensating for individual chip limitations [112]

Core Insights - The competition in artificial intelligence (AI) fundamentally revolves around computing power, algorithms, and data, with computing infrastructure being the cornerstone of this competition [1][2] - Unisplendour Corporation has successfully positioned itself among the top tier in the domestic market through both organic growth and external expansion [1] Group 1: AI Infrastructure and Competitive Edge - Unisplendour's unique feature in the AI infrastructure sector is its "integrated" ecosystem, leveraging its leading position in the network domain and rapidly growing computing product line [1] - The company possesses a rare capability for "computing-network collaborative design," enabling system architecture-level optimization of AI clusters, which reduces communication latency and enhances overall computing efficiency [1] Group 2: Strategic Positioning and Industry Trends - AI is penetrating various industries, evolving from general large models to industry-specific large models, positioning Unisplendour favorably in the wave of industrial intelligence [1] - The company's full-stack layout, encompassing cloud, network, security, computing, storage, and edge, allows it to provide comprehensive solutions in smart city projects, acting as a digital transformation partner rather than just a product supplier [1] Group 3: R&D and Market Position - In the context of global technological competition and the emphasis on self-reliance in China, the importance of a self-controllable ICT industry chain is increasingly highlighted [2] - Unisplendour aligns its development with the strategic direction of China's digital economy, continuously increasing R&D investment in high-end network equipment, key server technologies, and network security [2] - The company's experience illustrates that Chinese tech firms are actively competing for future discourse power in the global AI industry wave by focusing on core technologies and system-level innovations [2] Group 4: Competitive Challenges - Unisplendour faces comprehensive competition from top tech companies both domestically and internationally, with its ability to continue succeeding dependent on several key factors [2] - These factors include the depth and speed of technological innovation, breadth of breakthroughs in overseas markets, openness in ecosystem construction, and ongoing optimization of operational efficiency [2]

Core Insights - The company is strategically positioning itself for the AI era by sacrificing short-term profits for long-term value accumulation, evidenced by a 31.41% year-on-year revenue growth to 77.32 billion yuan in the first three quarters of 2025 [1] Strategic Anchors - The strategic acquisition of Xinhua San in 2016 marked a pivotal transformation for the company, involving nearly 35 billion yuan in investment over ten years, enabling entry into the AI computing power sector [2] Transformation and Competitive Advantage - Post-acquisition, the company has evolved from a network equipment specialist to a full-stack solution provider for the AI era, mastering the entire "cloud-network-security-computing-storage-end" chain, which is rare in the domestic digital infrastructure sector [3] Time for Space - The current phenomenon of "increasing revenue without increasing profit" reflects the company's strategic investment phase, with a net profit margin dropping to 2.23% despite significant revenue growth [4] Market Share Strategy - In the critical window of the AI computing arms race, prioritizing market share over short-term profitability is key, with low-margin customized AI server orders helping the company integrate into the mainstream AI ecosystem [5] White Box Opportunities - The rapid growth of the customized AI server market presents unique opportunities, akin to a "white box market," characterized by limited single-item profit margins but substantial and fast-growing market size [6] Scale as a Competitive Moat - Scale itself serves as a competitive moat in the computing infrastructure sector, with increasing demand for customized computing solutions as AI applications penetrate various industries [7] Global Competitiveness - The company's international performance validates its global competitiveness, with Xinhua San's international business revenue soaring by 60.25% year-on-year in the first half of 2025, significantly outpacing overall revenue growth [9] Technical Reserves - A key support for the company's long-term value is its substantial technical reserves, with a significant portion of revenue continuously invested in R&D to establish technological advantages in critical areas [11] Ecological Synergy - The company's potential lies in its industrial ecological synergy system, enabling it to provide end-to-end solutions from hardware infrastructure to software platforms and industry applications [13] Future Release - The company is transitioning from an investment phase to a harvest phase, with the demand for efficient, reliable, and green computing solutions expected to grow rapidly, benefiting firms with deep technological accumulation [15]

Summary of Key Points Core Viewpoint - The company has approved a guarantee for its wholly-owned subsidiary, Beijing Unis Digital Technology Co., Ltd., to secure a bank credit line of up to RMB 583.14 million for project financing [2][4]. Group 1: Guarantee Overview - The company will provide a guarantee for a bank credit line of RMB 583.14 million for its subsidiary, with the guarantee valid for 12 months from the date of the shareholder meeting [2]. - The guarantee is intended for the construction of the Beijing Changping Unis Digital Economy Technology Park (Phase I) [4][5]. Group 2: Guarantee Details - The loan amount is RMB 583.14 million, with a repayment period extending until January 19, 2040 [2][3]. - The guarantee includes coverage for principal, interest, penalties, and costs associated with debt recovery [4]. Group 3: Financial Implications - The total amount of guarantees provided by the company and its subsidiaries is RMB 1,774.71 million and USD 251.2 million, which represents 264.87% of the company's equity attributable to shareholders as of the end of 2024 [6]. - The current balance of guarantees to subsidiaries is RMB 1,337.31 million, which is 100.31% of the company's equity [6].

（编辑 袁冠琳） 证券日报网讯 1月23日，紫光股份发布公告称，截至目前，公司及子公司的担保额度总金额为人民币 1，774，714万元及251，200万美元，占公司2024年年末审计后的归属于母公司所有者权益的 264.87%。截至目前，公司及子公司对外担保余额为人民币1，563，314万元及250，700万美元，占公 司2024年年末审计后的归属于母公司所有者权益的248.75%。其中，对子公司的银行授信担保余额为人 民币1，337，314万元，占公司2024年年末审计后的归属于母公司所有者权益的100.31%；对子公司的 厂商授信担保余额为人民币226，000万元及248，000万美元，占公司2024年年末审计后的归属于母公司 所有者权益的147.03%；对合并报表外单位提供的担保余额为2，700万美元，占公司2024年末审计后的 归属于母公司所有者权益的1.42%。公司及子公司无逾期担保，无涉及诉讼的担保及因担保被判决败诉 而应承担的损失。 ...

Group 1 - The core viewpoint of the report emphasizes that the demand for AI computing power will continue to grow and diversify in 2026, extending from data centers to network edges and even internal terminals [8][13][27] - The report highlights the significant capital expenditure (Capex) growth driven by business revenue, with major cloud service providers like Google, Microsoft, and Amazon showing consistent revenue growth exceeding 20% year-on-year [27][31][35] - The report identifies the increasing importance of silicon photonics technology, particularly the 1.6T optical module, which is expected to maintain accelerated growth and enhance the global market share of domestic optical chips and devices [7][54][70] Group 2 - The report discusses the emergence of new technologies such as Scale-UP supernodes and their impact on optical link demand, indicating that these technologies will drive additional link requirements in 2026 [7][9][66] - It notes that the AI infrastructure is experiencing a high level of prosperity, with significant investments in data centers and AI capabilities expected to continue, particularly in the U.S. [47][49][53] - The report anticipates that the integration of silicon photonics will significantly increase its market penetration, projecting that by 2026, over half of optical module sales will come from silicon photonics solutions [70][75]

股票简称：紫光股份 股票代码：000938 公告编号：2026－004 紫光股份有限公司 关于为子公司提供担保的进展公告 本公司及董事会全体成员保证信息披露的内容真实、准确和完整，没有虚假记 载、误导性陈述或重大遗漏。 一、担保情况概述 紫光股份有限公司（以下简称"公司"）2025 年 12 月 3 日召开的第九届董事会 第十七次会议及 2025 年 12 月 19 日召开的 2025 年第六次临时股东会审议通过了《关 于为子公司申请银行综合授信额度提供担保的议案》，同意公司为全资子公司北京紫 光数字科技有限公司（以下简称"紫光数科"）自公司 2025 年第六次临时股东会审 议通过之日起12个月内申请的银行综合授信额度提供总额不超过人民币58,314万元 的担保（以下简称"本次额度"）。具体内容详见公司于 2025 年 12 月 4 日在巨潮资 讯网等披露的《第九届董事会第十七次会议决议公告》（公告编号：2025-066）、《关 于为子公司申请银行综合授信额度提供担保的公告》（公告编号：2025-068）及于 2025 年 12 月 20 日披露的《2025 年第六次临时股东会决议公告》（公告编号：2025- ...

紫光股份公告称，公司此前同意为全资子公司紫光数科申请的银行综合授信额度提供不超5.83亿元担 保。近日，公司与兴业银行北京分行签署《保证合同》，为紫光数科5.83亿元贷款提供连带责任保证， 贷款期限至2040年。本次担保后，公司在额度内对紫光数科担保余额为5.83亿元，截至目前，对其担保 余额为10.13亿元。截至目前，公司及子公司担保额度总金额为177.47亿元及2.51亿美元，对外担保余额 为156.33亿元及2.51亿美元，无逾期担保。 ...

Group 1 - The global share of China's essential patents for 5G standards has reached 42%, and the first phase of 6G technology trials has been completed, with over 300 key technology reserves formed [2] - The development of 6G technology is entering a critical window, with significant focus on the integration of low-orbit satellite networks and a comprehensive space-ground-sea network [2] - Companies are actively engaging in 6G-related business, with a focus on millimeter-wave technology as a core frequency band for 6G [3] Group 2 - Companies like 创远信科 are upgrading testing technologies and instruments towards higher frequency bands and greater bandwidth, with a focus on terahertz testing and AI communication testing [3] - 昂瑞微 is targeting the U6G frequency band and developing high-performance components, including high-power amplifiers and low-noise amplifiers, to enhance their competitive edge in RF solutions [4] - 紫光股份 is leveraging its strengths in ICT infrastructure to provide comprehensive ICT services for 6G networks, participating in various hardware and software service projects [4] Group 3 - 中兴通讯 has been actively involved in the IMT 2030 (6G) promotion group, contributing to testing and research in key technology areas such as 6G and AI integration [4] - 创远信科 has initiated the second phase of 6G technology trials and is focusing on the application of 6G+AI technology in specific scenarios like satellite internet and smart connected vehicles [5] - 瑞可达 is developing solutions for 6G communication and commercial satellite applications, while 臻镭科技 is positioning itself as a core chip supplier for 6G satellite internet [5]