各位线上的投资者朋友们大家下午好我是张旭津的策略分析师汪洋非常高兴在舍连春节的第一周和大家再次相聚在我们互联网的平台上和大家一起分享对于当前市场的一些观点 相关的APP上面应该可以看到的有RE和一个互联网相关的两个按钮然后我们可以看到RE的版本它的性能直接对标了OpenAI的 OE的正式版本然后它在数学在代码在自然语言推理等各项任务上面它都能够比肩OpenAI的这样的一个最新的版本然后另外我们会看到15年1月份到春节前到春节这段时间DeepSeek的APP的上线的20天的日活的用户数都超过了2000万人次 然后在全球的一些应用商店来讲的话软件的应用商店来讲的话也多次位列全球软件的下载榜的榜首现在我们也看到有很多的一些国外的大型科技公司已经先后上线部署了知识用户去访问DeepSafe R1的模型然后不管是国内的还是海外的一些顶尖的一些科技融合公司 都已经开始在使用这个DeepSync相关的RE的这样一些模型那我们了解了这个大概的背景之后呢我们再来看一下就是DeepSync它在性能方面在价格方面在开源方面 它对于整个的人工智能模型方面是带来了什么样的一些冲击我给大家一些数据供大家来做参考首先DeepStack R ...

市场对于这个AI算力的这个长期需求的判断其实产生了一定分歧的那么最近的我们也跟国精的这个科技团队的这个分析实验有老师啊就密集的做了一个沟通那么得到一个这个比较核心的一个基本结论就是啊随着这个应用端的一个加速爆发那么对整体的这个算力需求的一个预测啊应该说比较大概率还是一个有增 那么这个最近几家大家看到的这个发布财报的科技巨头的这个资本开支啊也是一个继续维持的一个高增速的这么一个状态所以目前基本上可以判断就是在未来比较长的一段时间里面啊算力相关的这个基础设施的建设和投资那么很重要一块就是这个电力供应这一块那么知道啊应该说不会有这个步伐放缓的这个情况啊包括 电力的供应和这个数据中心相关的一个供配电的一个设备所以就是这个落实到投资这块来看的话A股的这个相关的这个AI电力相关的标的最近也确实因为这种对于算力需求判断的这个分析 本次电话会议仅面向国金证券的专业投资机构客户或社交客户仅供交流研究观点专家发言内容仅代表其个人观点会议内容并不构成对任何人的投资建议未经国金证券 好的各位投资者大家晚上好欢迎大家在周五的晚上来介入我们国庆电信团队 近期发布的一篇关于AIDC供配电系统的一个专题深度报告的一个解读电话会我是国庆电信的 ...

更多资料加入知识星球：水木调研纪要 关注公众号：水木纪要 AI 如何赋能 SaaS 重估值 摘要 • AI 应用中令牌成本显著下降（80%-90%）将大幅提升相关企业净利率，弹 性或达 30%，尤其利好高比例依赖令牌成本的公司，带来基本面实质性改 善，而非仅是题材炒作。 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 Q&A 近年来，A 股的云计算公司纷纷宣布接入 DEX SAP RY 模型，这对公司的影响如 何？ A 股典型的云计算公司如金蝶、用友、金山办公等纷纷宣布接入 DEX SAP RY 模 型进行适 ...

国经济都有负面影响。因此，台积电在美国加速建厂可能是双方平衡点。 大摩内部会：从 DeepSeek 看 AI 供应链新浪潮 摘要 Q&A 从半导体行业的角度来看，AI 供应链有哪些转变和新的浪潮？投资机会如何？ 首先，端侧的大模型正在变小，这是 AI 发展的一个必要条件，但还需观察具体 应用的推动。我们关注 MediaTek 的业绩报告，因为政府对 DPC 设备的需求较高， 这为基本算力提供了支持。英特尔在春节期间推出的新产品有优势，因为它对 冷却系统依赖较低，因此我们推荐 Lattice Semiconductor（LTC）。英伟达及 其合作伙伴台积电在短中长期内都有影响，我们需要密切关注这些公司的发展。 近期，专家会议讨论了新发明及其低成本对硬件的影响。大公司 CEO 普遍认为 双 PM（生产管理）仍然重要，就像酿造优质葡萄酒需要好的葡萄一样。因此， PSP（生产服务提供商）对大规模投资持乐观态度，目前恐慌情绪已过去。在大 中华区上游市场，如台积电及其周边企业，包括月光、天外、上泉和万润等设 备供应商，现在是一个好的买点。 短期内，MEV 汽车组装不顺利导致出货量下 降，但这主要是供应问题而非需求问题。下 ...

Summary of the Conference Call on Greater China Technology Semiconductors Industry Overview - The report focuses on the **Greater China Technology Semiconductors** industry, specifically analyzing the performance of Taiwan's tech companies in December 2024 [1][2]. Key Financial Performance - **December Revenue**: Taiwan's tech companies experienced a **1% decrease month-over-month (MoM)** but a **35% increase year-over-year (YoY)**. Notably, several AI server supply chain companies achieved record monthly revenues [4][2]. - **Performance by Segment**: - **Datacenter/Smartphone Hardware and Packaging & Testing** companies exceeded expectations. - **Virtual Reality (VR), semiconductor materials, and thermal solutions** underperformed relative to expectations [4][2]. Investment Preferences - **Preferred Stocks**: The report recommends stocks with strong order visibility or leading market positions, including: - TSMC - KYEC - AllRing - ASMedia - Hon Hai - AVC - King Slide - E-Ink - Accton - Wiwynn [4][2]. - **Underweight Recommendations**: The report identifies Nanya Technology, Macronix, and NYPCB as key underweight (UW) stocks [4][2]. Market Outlook - The overall industry view is classified as **In-Line**, indicating that the performance of the semiconductor sector is expected to align with broader market trends over the next 12-18 months [2][4]. Additional Insights - The report emphasizes the dominance of **Artificial Intelligence (AI)** in driving revenue growth within the semiconductor sector, highlighting its critical role in shaping future market dynamics [1][2]. - Analysts express a cautious outlook on segments that did not meet expectations, suggesting potential risks for investors focusing on those areas [4][2]. Conclusion - The December revenue wrap-up indicates a mixed performance across Taiwan's tech companies, with AI-related sectors showing strength while others lagged. Investment strategies should focus on companies with robust order visibility and market leadership to navigate the current landscape effectively [4][2].

Summary of J.P. Morgan's 2025 Internet Sector Survey Industry Overview - **Industry**: Internet Sector - **Date**: January 12, 2025 - **Survey Participants**: Nearly 100 investors Key Findings Market Performance Expectations - **US Internet Sector**: ~79% of respondents expect market-cap weighted Internets to rise more than 5% in 2025, with 38% anticipating increases of 15% or more [2][5] - **S&P 500**: Only ~66% expect the S&P 500 to increase by more than 5% [2][5] Economic Outlook - **Recession Probability**: ~76% of respondents believe there is a 25% or less chance of a recession in 2025 or the first half of 2026 [2][5] Tailwinds for the Internet Sector 1. **Advances in Technology & Innovation/New Products**: 17% [7][25] 2. **Return on AI Investments**: 14% [7][25] 3. **Upward Earnings Revisions**: 13% [7][25] Headwinds for the Internet Sector 1. **Lack of AI Monetization**: 19% [7][30] 2. **Heavy Capex/Margin Compression**: 17% [7][30] 3. **Unattractive Valuation**: 17% [7][30] Generative AI Expectations - **New Applications**: 37% expect compelling new GenAI agents and applications [2][36] - **Revenue Contribution**: Only ~6% expect material revenue contributions from GenAI/LLMs [2][36] Stock Performance Predictions - **Best Performing Internet Stock**: GOOG/L is expected to be the best performer among Internet Mag 7 stocks with 43% of respondents [2][46] - **Worst Performing Internet Stock**: META is anticipated to be the worst performer with 46% of respondents [2][49] - **Best Performing Large-Cap Stock**: UBER is expected to lead with 40% [2][56] - **Best Performing Mid-Cap Stock**: RDDT is expected to lead with 22% [2][61] - **Best Performing Small-Cap Stock**: Z/ZG is expected to lead with 18% [2][67] - **Best Performing Sub-Sector**: Subscription-Based Companies (26%) and Online Ads (23%) are expected to perform best [2][77] Investor Preferences - **Rideshare/Food Delivery**: UBER is preferred by 51% of respondents [2][81] - **Online Travel**: BKNG is favored by 38% [2][81] - **Video Games**: RBLX and TTWO are both favored by 28% [2][81] Turnaround Stories - **Best Turnaround Stories**: UBER (31%), PINS (16%), and ABNB (15%) are viewed as the best turnaround stories for 2025 [2][81] Additional Insights - **Investor Composition**: 55 Long Only, 30 Hedge Fund, and 5 Others participated in the survey [2][5] - **Survey Timing**: Conducted from January 3 to January 10, 2025 [2][12] This comprehensive survey provides a positive outlook for the Internet sector in 2025, with significant expectations for growth driven by technological advancements and AI investments, despite some concerns regarding valuation and profitability.

Summary of Conference Call Notes Industry Overview - The focus is on the AI infrastructure and its investment strategies, particularly in the context of the "二五联盟" (2025 Alliance) and its implications for the AI sector [1][2] - The AI infrastructure is expected to experience a positive growth trend, distinguishing itself from previous years [1] Key Insights and Arguments - AI infrastructure is projected to be in a phase of sequential growth, with significant advancements expected in AI applications and technologies [1] - The introduction of AI agents (Agin) is highlighted as a core driver for future growth, with recent product launches from TechGpt indicating a shift towards more advanced AI models [1] - The performance of various sectors since the launch of the AI infrastructure two years ago shows that energy and database security software services have seen substantial growth, outperforming major indices like NASDAQ and S&P [3] - The report categorizes AI's impact into three stages: product launch, positive progress, and actual contribution to revenue, with infrastructure companies showing higher AI realization compared to software and application firms [4] Financial Performance - Revenue growth rates are showing a declining trend, indicating a need for companies to focus on actual earnings per share (EPS) and order catalysts [4] - The cumulative growth of AI applications since the launch has been 115%, with significant variances in performance across different sectors [3] Sector-Specific Recommendations - **Semiconductors**: Emphasis on monitoring chip shipment volumes and the competitive landscape among GPU manufacturers [5][7] - **Cloud Computing**: Focus on the ROI of cloud vendors in the AI era and the evolving competitive dynamics [5] - **Software Services**: Companies that have launched AI tools and are seeing clear customer adoption should be prioritized for investment [6][12] - **Energy Sector**: Data centers are expected to drive demand for electrical components and thermal management systems [5][8] Emerging Trends - The AI landscape is evolving with new product launches in AI PCs, smartphones, and glasses, indicating a broader market penetration [9][10] - The second year of AI applications has seen a shift in market focus from hype-driven growth to actual performance metrics, with a notable decline in the performance of B2B software [10][14] - The education and gaming sectors are leveraging AI to enhance user engagement and subscription rates, indicating a growing trend in AI adoption across various industries [15] Future Outlook - The overall sentiment is optimistic about the AI sector's potential to return to accelerated growth, with a focus on companies that can demonstrate actual contributions from AI initiatives [12][16] - The importance of monitoring the performance of AI tools and their impact on revenue generation is emphasized, particularly in the context of B2B software [12][22] - The report concludes with a call to focus on AI's contributions to infrastructure and applications as key drivers for investment strategies leading up to 2025 [23][24]

Summary of Key Points from Conference Call Company and Industry Overview - **Company**: NVIDIA - **Industry**: Artificial Intelligence (AI), Robotics, Semiconductor Core Insights and Arguments 1. **Project Digit AI Box Applications**: Project Digit AI is designed for edge AI processing, focusing on lightweight pre-training models, fine-tuning, and inference tasks. It features a high level of integration with NVIDIA's proprietary architecture, including Grace CPU and GB series GPU versions, and operates as a closed-edge training device without third-party dependencies [1][2][3] 2. **Market Positioning of Project Digit**: Targeting small to medium-sized B2B users, Project Digit fills a gap in NVIDIA's traditional data center offerings. It is tailored for professional users like SMEs, developers, and startups, providing low deployment costs and ease of use, while lacking appeal for end consumers [2][3] 3. **Comparison with CSP Services**: Project Digit's low deployment cost and plug-and-play features attract small B2B customers who may not have extensive model development capabilities. However, customers seeking flexibility may prefer CSP services, which offer a wider range of model options and better scalability [3] 4. **GB200 NVL72 Design Features**: The GB200 NVL72 is a next-generation design for large-scale computing clusters, achieving higher integration density through wafer-level interconnects. It is still in the early stages of development and faces challenges in general data center acceptance due to its stringent design requirements [4] 5. **Cerebras and OpenAI Collaboration**: Cerebras is collaborating with OpenAI to run OpenAI's models on Cerebras devices, providing an alternative to traditional GPU setups [5] 6. **Impact of Test Time Scaling on Inference Market**: NVIDIA holds over 95% of the training market but aims to enhance its GPU's role in inference through the test time scaling principle, ensuring it maintains and expands its market share amid the growth of large models [6] 7. **Cosmos World Model and Tesla FSD**: The Cosmos model currently does not directly compete with Tesla's FSD due to Tesla's extensive road testing data. However, it offers automakers a lower barrier to entry into autonomous driving, promoting a decoupled hardware-software approach [7] 8. **NVIDIA's Strategy in Robotics**: NVIDIA is entering the humanoid robot market, recognizing its potential despite Tesla's current lead. The company aims to maintain a competitive yet non-conflictual relationship with Tesla, emphasizing solution-oriented approaches [8] 9. **Tesla's Production Goals for Optimus Robots**: Tesla's target of producing 500,000 Optimus robots in the next three years is deemed feasible due to its comprehensive supply chain capabilities [9] 10. **OpenAI's Robotics Progress**: OpenAI plans to announce its robotics products in 2025, focusing on developing complete end-to-end solutions in collaboration with third-party partners [10] 11. **HP's Success with x.AI Orders**: HP secured a $1 billion order from x.AI due to its deep customization capabilities and close collaboration during the design phase, contrasting with competitors like Dell and Supermicro [11] 12. **HP's Liquid Cooling Delivery Capability**: HP has developed its own liquid cooling solution, demonstrating comparable performance to competitors and showing readiness for large-scale delivery [12] 13. **OpenAI's ASIC Development with Broadcom**: OpenAI is working on ASICs to replace some NVIDIA GPUs, with plans for small-scale internal deployment in Q4 2023 and larger production expected by 2026 [13] 14. **Apple and Broadcom's Baltra AI Chip**: The Baltra chip focuses on inference tasks, complementing Apple's existing training resources and expected to be operational by early next year [14] 15. **Meta's MTIA Orders with Broadcom**: Meta's MTIA has two versions, with the first underperforming and the second expected to ramp up production in 2026, indicating continued reliance on NVIDIA products [15] 16. **ZJ ASIC Project Status**: The ZJ ASIC project is facing delays due to policy and international environment issues, with completion now expected in 2026 [16] 17. **ZJ's Orders for HWJ and Shenteng**: ZJ's orders for HWJ focus on advertising search model training, while Shenteng's orders pertain to traditional AI applications [17] 18. **HWJ's Shipment Expectations**: HWJ is prioritizing orders from ByteDance, with production capacity largely dependent on SMIC's yield improvements [18] 19. **HWJ590 Yield Improvement**: The HWJ590 model has seen yield improvements recently, although it initially faced challenges due to SMIC's production quality [19] 20. **690 Chip Production Progress**: The 690 chip is currently in design, with early production at TSMC expected to yield small batches post-Chinese New Year [20] 21. **Microsoft's Investment Considerations in Anthropic**: Microsoft is exploring a partnership with Anthropic due to dissatisfaction with OpenAI, which has become competitive with its Copilot product [21] 22. **Trainium 3 Development**: Trainium 3 is being developed by Alchip, with Marvell also vying for the project [22]

Key Points Industry Overview 1. **Domestic Copper Foil Market Supply**: Limited supply, accounting for less than 10% of overall demand. Main suppliers include Shandong Glass Fiber and Honghu, but overall numbers are low. Over 90% of high-end copper foil materials like electronic preservation foil, Sumitomo, HVRP copper foil, and Sabic resin rely on imports, mainly from Japan and Taiwan. [3] 2. **AI Server Market**: Expected to reach 4 billion RMB in PCB orders in 2025, but domestic manufacturers face challenges due to insufficient capacity and reliance on imported high-end materials. Market share is expected to increase to around 30%. [4] 3. **Copper Foil Prices and Supply**: Currently high and tight, with Taiwan manufacturers raising prices by controlling supply. Prices range from 100 to 120 RMB for a standard copper foil to 500 RMB for high-speed copper clad laminates. [14] Key Players and Suppliers 4. **Major Material Suppliers in AI Server Industry**: Taiwan's Chimei, South Korea's Doosan, and Taiwan's Lianhai are recognized by NVIDIA. Chinese mainland companies like Shengye Electronics, Shanghai Nanya Electronics, and华盛电子 have a smaller market share and are actively trying to enter NVIDIA's supply chain. [5] 5. **NVIDIA Copper Foil Supplier Decision**: Determined by NVIDIA's CCL (Copper Clad Laminate) factory, not NVIDIA directly. NVIDIA focuses on final product performance indicators such as glass strength, wave resistance, and impedance. [6] 6. **Progress of Chinese Mainland Companies in Entering NVIDIA Supply Chain**: Several Chinese mainland companies are trying to enter NVIDIA's supply chain, including Jiangsu Garden Copper Foil, Defu, and Yihao. They have started sampling and delivering PCB boards to NVIDIA for testing. [7] Market Dynamics and Challenges 7. **Application Scenarios of Low-Level HVL-1 Copper Foil**: Suitable for cost-sensitive applications, not mainstream high-end applications. Used in non-core components or older models, while high-end devices like 5,090 GB200 servers or GB H100 require higher-level materials like HVL-2 or HVL-3. [8] 8. **Material Utilization Rate and Improvement**: High-end press machines with fixed dimensions result in low copper foil material utilization rate (about 70%). Improvement requires focusing on other process links rather than relying on existing press machines. [13] 9. **Domestic RTF-2 Copper Foil Supply**: Main suppliers include Yihao, Defu, and Jiangnan Copper. Overall supply占比 only 10% to 20%, and Taiwan manufacturers have limited supply to the mainland market. [22] Future Outlook 10. **Market Share of Domestic AI Servers**: Expected to reach around 30%. Domestic companies like DeepSouth Circuits and Guanghe Business Vibration are actively preparing to produce domestic AI servers and gradually improve their performance and quality. [12] 11. **Challenges for Domestic Replacement**: Some high-end materials still cannot meet demand, such as M7 and M8 levels. Comprehensive domestic replacement requires time and technical accumulation. [27]

Summary of Conference Call Company/Industry Involved - The discussion primarily revolves around the **Artificial Intelligence (AI)** sector and its investment opportunities within the **A-share market** in China. Core Points and Arguments 1. **AI Market Growth in 2024**: The AI sector experienced significant growth in 2024, with advancements in multimodal models like OpenAI's SORA, which expanded capabilities from text to video [2][3]. 2. **Investment Focus on AI**: The investment strategy has been heavily focused on AI, particularly in computing power, which is expected to continue driving growth in the sector [4][5]. 3. **Performance of AI-related Stocks**: AI-related stocks, especially those linked to computing power, showed strong performance throughout 2024, reflecting the industry's overall positive sentiment [3][4]. 4. **Market Volatility and Strategy**: The market faced volatility, particularly after September 2024, but the strategy of focusing on high-growth AI stocks proved effective, leading to a recovery in December [9][10]. 5. **2025 Outlook for AI**: The outlook for 2025 remains optimistic, with expectations of continued growth in AI applications and infrastructure, despite concerns about potential slowdowns in demand for computing power [16][19]. 6. **Investment Opportunities in AI Applications**: There is a belief that significant innovation in AI applications, particularly in B2B and B2C sectors, will emerge in 2025, creating new investment opportunities [20][21]. 7. **Competitive Advantage in China**: China's AI sector is expected to leverage its large consumer base and rapid iteration in applications, potentially leading to a global leadership position in AI applications [23][24]. 8. **Valuation and Performance of AI Stocks**: Current valuations of AI infrastructure stocks are considered reasonable, with no signs of a bubble, while application-focused companies may see valuation increases as innovations emerge [25][27]. 9. **Risks in AI Investment**: Key risks include supply chain issues related to GPUs and chip shortages, which could impact the overall AI industry [30]. Other Important but Possibly Overlooked Content 1. **Investment Strategy Framework**: The investment approach is based on a comprehensive research framework that integrates insights across technology, consumer, and cyclical sectors, allowing for a more holistic view of the market [12][13]. 2. **Historical Context**: The discussion draws parallels between the current AI landscape and past technological revolutions, suggesting that the AI sector may follow a similar trajectory of rapid growth and innovation [19][23]. 3. **A-share Market Sentiment**: The overall sentiment towards the A-share market is optimistic, with expectations of improved economic conditions and increased capital support from the government [31][32]. This summary encapsulates the key insights and strategic outlook discussed during the conference call, highlighting the potential and challenges within the AI sector and the broader A-share market.