创业板人工智能ETF华夏（159381）跟踪指数CPO权重近50%，同时覆盖国产软件+AI应用企业，具备 较高弹性。其中前三大权重股为中际旭创（15.64%）、新易盛（15.57%）、天孚通信（6.85%）。场内 综合费率仅0.20%，位居同类最低。场外联接（A类：025505；C类：025506）； 通信ETF华夏（515050）深度聚焦电子（芯片、PCB、消费电子）+通信（光模块、服务器）算力硬 件。前五大持仓股：中际旭创、新易盛、立讯精密、工业富联、兆易创新。场外联接(A类：008086；C 类：008087) 中信证券研报称，近期，海外推理和训练算力需求旺盛，亚马逊云和谷歌云双双涨价。推理侧， MoltBot（原ClawdBot）、Claude Code等Agent产品加速落地，对于云计算资源的需求显著增加，2026 年年初以来Token调用量已经连续2—3周高速增长。训练侧，Grok-5、Veo4等模型仍然在持续迭代，工 业界持续探索Scaling上限，从而支撑训练算力需求。尽管当前AI应用大规模商业化落地能见度仍然不 足，但展望未来3—6个月，随着推理端AI应用落地的密集催化叠加训练端模型的持续迭 ...

1月30日，A股三大指数早盘涨跌不一，截至午间休市，沪指跌1.19%，深证成指跌0.96%，创业板指涨 0.8%。沪深北三市半日成交19514亿元，较上日缩量836亿元。 盘面上，亚太交易时段国际金银价跳水，黄金股、有色资源股大幅回调，中金黄金等多股跌停。另一方 面，粮食、农业股逆势走强，登海种业等股封板。算力硬件股盘中同样拉升，天孚通信（300394）大涨 超10%创出历史新高，中际旭创、新易盛涨超4%。 行业层面，近期，海外推理和训练算力需求旺盛，亚马逊云和谷歌云双双涨价。推理侧， MoltBot（原 ClawdBot）、Claude Code等Agent产品加速落地，对于云计算资源的需求显著增加，2026年年初以来 Token调用量已经连续2-3周高速增长。训练侧，Grok-5、Veo4等模型仍然在持续迭代，工业界持续探索 Scaling上限，从而支撑训练算力需求。 中信证券研报认为，尽管当前AI应用大规模商业化落地能见度仍然不足，但展望未来3-6个月，随着推 理端AI应用落地的密集催化叠加训练端模型的持续迭代，预判算力需求仍有望进一步上行，一旦美股 财报季算力需求得到确认，前期压制算力板块情绪与估值的 ...

Group 1 - The core viewpoint of the news highlights a significant rebound in computing hardware stocks, with notable increases in share prices for several companies [1] - LianTe Technology experienced a nearly 10% increase in stock price, reaching a historical high [1] - Tianfu Communication and Changfei Optical Fiber both saw stock price increases exceeding 8%, also achieving historical highs [1] Group 2 - Other companies such as Shijia Photon, Hengtong Optic-Electric, Dekoli, and Xinyi Sheng also experienced stock price increases, indicating a broader positive trend in the sector [1]

近期披露完毕的2025年公募基金四季报显示，中际旭创（300308）、宁德时代（300750）、新易盛 （300502）、紫金矿业（601899）、寒武纪-U、立讯精密（002475）、贵州茅台（600519）、东山精 密（002384）、美的集团（000333）、中国平安（601318）成为主动基金的前十大重仓股。 纵观过去的公募基金重仓股名单，不难发现市场的热点赛道在不断变化。2024年三季度前，贵州茅台长 期稳坐公募基金重仓股第一的"宝座"。之后，随着新能源产业的崛起，宁德时代取代了贵州茅台，并连 续4个季度成为公募基金第一重仓股。直到现在，AI行业兴起，中际旭创成为基金重仓股。可以预见的 是，以AI算力基础设施、AI芯片、AI应用等领域为代表的AI相关产业链将继续成为未来一段时间内市 场关注的焦点。但与此同时，在科技成长风格持续主导的背景下，投资者需要关注相关行业短期估值水 平和业绩增长情况，避免过度追高。 从重仓股排行可以看出，主动权益类公募基金的持仓结构发生了明显变化，AI等科技成长股成为"重点 关注对象"。事实上，科技行业上市公司比重增加不仅体现在公募基金前10大重仓股上，也体现在整体 的持仓变 ...

寒武纪-U是一家AI芯片公司，专注于云端和边缘计算的人工智能芯片研发。作为国内AI芯片领域 的领先企业，其股价此前曾一举超过贵州茅台成为A股"股王"。早在2024年四季度，寒武纪-U就跻身基 金重仓股第8位，之后除了在2025年三季度掉出前20名外，一直在10名左右徘徊。在AI应用的快速发展 带动下，寒武纪-U在2025年四季度一举升至基金重仓股第5位。 从重仓股排行可以看出，主动权益类公募基金的持仓结构发生了明显变化，AI等科技成长股成 为"重点关注对象"。事实上，科技行业上市公司比重增加不仅体现在公募基金前10大重仓股上，也体现 在整体的持仓变化上。Wind数据显示，公募基金2025年四季度对科技行业上市公司的持仓已经从2024 年一季度的25%上升到40%。与此同时，新能源、消费、金融行业的占比有所下降。这一变化反映了市 场对科技行业特别是AI行业发展的高度关注。 近期披露完毕的2025年公募基金四季报显示，中际旭创、宁德时代、新易盛、紫金矿业、寒武纪- U、立讯精密、贵州茅台、东山精密、美的集团、中国平安成为主动基金的前十大重仓股。 在人工智能（AI）产业的投资浪潮下，科技成长风格成为主导。其中，中际 ...

Group 1 - The core viewpoint of the articles highlights the increasing demand for AI-driven cloud computing services, leading to a tightening supply-demand dynamic in the computing power sector, which is reflected in rising prices from major cloud providers [8][11]. - The "算力ETF" (Computing Power ETF) has seen significant inflows, with over 1.4 billion yuan accumulated in the last ten days, indicating strong investor interest in the sector [1]. - Major tech companies like Microsoft, Meta, and others have reported impressive cloud revenue, with Microsoft achieving $50 billion in cloud revenue for the second fiscal quarter, surpassing analyst expectations [4]. Group 2 - Clawdbot, an open-source AI assistant, has gained significant attention, with deployment support from Tencent Cloud and Alibaba Cloud, indicating a shift towards more personalized AI applications [3][6]. - The pricing strategies of leading cloud providers, such as Google and AWS, have been adjusted upwards, reflecting the increased costs associated with AI training and inference demands [8][9]. - The computing power supply chain is experiencing inflation, with projections indicating significant price increases for DRAM and NAND storage, as well as CPU products, driven by AI demand [9][10]. Group 3 - The rapid growth in token usage for AI models, particularly from OpenAI and Google, suggests a burgeoning market for AI applications across various sectors, with daily token calls reaching unprecedented levels [10]. - The anticipated rise in AI model applications is expected to further drive demand for computing resources, benefiting companies involved in hardware, cloud services, and AI development [11]. - The computing power ETF, 汇添富 (159273), is positioned to capture the growth opportunities in the AI-driven cloud computing market, covering a wide range of sectors including hardware, IT services, and data center operations [12].

Group 1 - The core viewpoint of the news highlights the performance of the AI-themed index and ETF, with notable gains in specific stocks like iFlytek and Kunlun Wanwei, while the AI ETF shows significant trading volume and growth in assets [1][2] - As of January 28, the AI-themed ETF has a total asset size of 34.06 billion yuan, with a recent increase of 2.24 million shares over the past six months, indicating strong investor interest [1][2] - The AI-themed index consists of 50 companies involved in providing resources, technology, and application support for artificial intelligence, with the top ten stocks accounting for 58.08% of the index [2] Group 2 - The storage chip supply tightness and price increases driven by the AI infrastructure boom are expected to persist until 2027, potentially leading to a systematic upcycle in the chip industry [2] - Domestic chip manufacturers are expected to raise prices, reflecting enhanced pricing power and accelerated localization, which may improve gross margins and cash flow, supporting R&D and expansion [2] - The AI ETF closely tracks the performance of the AI-themed index, which is designed to reflect the overall performance of listed companies in the AI sector [2]

Summary of Conference Call Notes Industry Overview - The conference call discusses the optical communication industry, specifically focusing on two companies: Eoptolink and ZhongjiInnoLight. Key Points - Eoptolink and ZhongjiInnoLight received regulatory guidance to delay their earnings forecasts compared to their original plans. This decision aims to facilitate a smoother and gradual upward trend in the market [1][2]. - The guidance provided by regulators is not due to concerns that either company may fail to meet profit expectations, but rather for the purpose of market rhythm control [2][3]. Additional Important Information - The regulatory intervention indicates a proactive approach to managing market conditions, suggesting that authorities are closely monitoring the performance and expectations of companies within the optical communication sector [1][2].

2026 年 01 月 28 日 华夏中证科创创业 50ETF 投资价值 分析 ——指数基金产品研究系列报告之二百六十七 相关研究 证券分析师 沈思逸 A0230521070001 shensy@swsresearch.com 邓虎 A0230520070003 denghu@swsresearch.com 联系人 沈思逸 A0230521070001 shensy@swsresearch.com 权 益 量 化 研 究 证 券 研 究 报 告 请务必仔细阅读正文之后的各项信息披露与声明 本研究报告仅通过邮件提供给 博时基金 博时基金管理有限公司(researchreport@bosera.com) 使用。1 ETP 研 究 - ⚫ 华夏中证科创创业 50ETF 于 2021 年 6 月 24 日成立，截至 2025 年末规模 50 亿元， 费率 0.15%，目前由徐猛和王欣薇两位基金经理共同管理。 ⚫ 当前，中国经济发展的核心驱动力正在发生深刻变革。推动产业转型升级、培育经济增长 新动能，关键在于以科技创新驱动新质生产力的发展。国家明确将统筹各方力量坚决打赢 关键核心技术攻坚战。特别是在面向未来的"十五五" ...

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]