7月8日，三大指数集体上涨，沪指收涨0.70%。通信ETF大涨4.65%，创业板人工智能ETF国泰也走出3.78%的涨幅。 数据来源：wind 除了GPU相关的需求旺盛之外，ASIC的需求也是水涨船高。最新的TPU在今年4月发布，谷歌已经推出了第七代张量处理单元（TPU）Ironwood。谷歌称， 在大规模部署的情况下，这款AI加速器的计算能力能达到全球最快超级计算机的24倍以上。在ASIC市场，目前博通以55%～60%的份额位居第一。博通的财 报也表明了ASIC业务成为了其增长的核心动能。目前博通有两个大合作备受关注：第一是Meta与博通已合作开发了前两代AI训练加速处理器，目前双方正 加速推进第三代MTIA芯片的研发。第二是OpenAI已委托博通开发两代ASIC芯片项目，计划于2026年投产，将采用业界领先的3nm/2nm制程工艺并搭配 3DSOIC先进封装技术。 消息面上，昨晚工业富联发布《2025年半年度业绩预告》，大幅超出市场预期。经初步测算，公司预计2025年第二季度归母净利润67.27亿元到69.27亿元， 同比上升47.72%到52.11%。同时，公司在预告中提到："AI服务器营业收入较去年同 ...

Market Overview - The market showed strong performance with all three major indices rising, particularly the ChiNext Index which surged by 2.25%, and over 3,900 stocks in the market increased in value, indicating heightened capital activity [1] Solar Energy Sector - The solar energy sector experienced significant gains with stocks like Shihang New Energy, Yijing Photovoltaic, and Tongwei Co. hitting the daily limit, driven by supportive policies from the Ministry of Industry and Information Technology aimed at promoting high-quality development in the solar industry [3] - New policies in regions like Yunnan are supporting the integration of solar manufacturing and power generation, leading to accelerated supply-side reforms and the exit of outdated production capacity, benefiting leading companies [3] - Technological innovations such as BC batteries and perovskite breakthroughs have greatly enhanced the competitiveness of solar companies [3] Computing Hardware Sector - The computing hardware sector saw notable performance with companies like Yihau New Materials and Xinyi Sheng reaching historical highs, driven by the explosive demand for AI computing power following the launch of NVIDIA's GB200 server [3] - The development of ASIC chips is accelerating, with expected shipments surpassing GPUs next year, which is a long-term positive for the entire computing supply chain [3] - Industrial Fulian, a leading global electronics manufacturer, is deeply tied to top clients like NVIDIA, showing strong performance in stock prices [3] Gaming Sector - The gaming sector also performed well, with companies like Ice Glacier Network and Giant Network seeing significant gains due to a substantial increase in the number of game licenses, reaching a 22-year high in June [4] - Supportive policies for the gaming industry are enhancing market expectations for profitability, while AI technology is improving development efficiency and user experience [4] Building Materials Sector - The building materials sector showed positive performance with companies like International Composite Materials and Shandong Fiberglass hitting the daily limit, driven by ongoing "anti-involution" policies aimed at improving industry fundamentals [4] - The central financial meeting emphasized the need to eliminate low-price disorderly competition, which is favorable for leading building material companies [4] Consumer Electronics and PCB Sectors - The consumer electronics sector benefited from increased tariffs on Vietnam by the U.S., highlighting the advantages of Chinese manufacturing, alongside domestic policies stimulating demand [4] - The PCB sector is experiencing growth due to surging demand for high-end products in AI servers and automotive electronics, with leading companies accelerating production [4] Underperforming Sectors - The banking sector saw declines, with major banks like Shanghai Pudong Development Bank dropping over 2%, attributed to tightening liquidity at month-end and institutional reallocation of funds [5] - The power sector also faced a pullback, influenced by changing demand expectations following high temperatures [5] - The insurance sector and controlled nuclear fusion sector underperformed due to regulatory impacts and previous overvaluation, respectively [5] Conclusion - Overall, the market displayed a clear structural trend with strong performances in solar, computing, and gaming sectors, while banking and power sectors lagged behind [5]

Core Viewpoint - In the first half of the year, 52 A-share companies reported their performance forecasts, with over 60% showing positive results, particularly in the electronics sector where companies like TaiLing Micro, XinDong LianKe, XinPeng Wei, and ChangChuan Technology saw median net profit growth exceeding 80%, with the highest reaching 267% [1] Group 1: TaiLing Micro (688591) - TaiLing Micro's performance surge is attributed to its innovative applications in AIoT, such as low-power Bluetooth chips for smart glasses, enhancing battery life from one charge per day to three [2] - The company secured significant orders from major retail chains by integrating multi-mode chips into electronic price tags, enabling seamless Bluetooth and Wi-Fi switching [2] - With local foundries prioritizing its production, TaiLing Micro's cost has decreased, and its gross margin has soared to 60% [2] Group 2: XinDong LianKe (688582) - XinDong LianKe has successfully penetrated the MEMS sensor market, traditionally dominated by foreign giants, by focusing on high-end applications like attitude sensors for commercial satellites and dynamic measurement for autonomous vehicles [4] - The company has received over 400 million yuan in orders for commercial aerospace and smart driving sectors in the first half of 2025, surpassing its total revenue for the previous year [4] - XinDong LianKe manages the entire process from chip design to packaging, ensuring production stability despite supply chain disruptions, achieving a gross margin of 85% [4] Group 3: XinPeng Wei (688508) - XinPeng Wei excels in the analog chip market, with its products integrated into various applications, including electric vehicle chargers and photovoltaic inverters [5][6] - The company reported a 70% revenue increase in industrial automation and automotive electronics sectors for the first half of 2025, driven by its versatile product strategy [6] - XinPeng Wei benefits from domestic chip policies, leading to a surge in orders for imported alternatives [6] Group 4: ChangChuan Technology (300604) - ChangChuan Technology has capitalized on the booming semiconductor testing equipment market, driven by the global demand for AI chips and HBM storage chips [8] - The company offers competitive pricing, with its testing machines being 30% cheaper than international counterparts while maintaining similar performance [8] - Following a successful fundraising of 3.1 billion yuan, ChangChuan Technology plans to double its production capacity for high-end testing machines, positioning itself strongly in the AI chip testing market [8] Group 5: Industry Insights - The success of these companies stems from their focus on niche markets like AIoT chips and MEMS sensors, allowing them to excel without directly competing with industry giants [9] - Many domestic firms are achieving performance levels comparable to international leaders while offering more attractive pricing [9] - Emerging sectors such as low-altitude economy, AI terminals, and electric vehicles are creating new growth opportunities in the semiconductor industry, benefiting early movers [9]

①特朗普签署"大而美"法案对各类投资影响几何？ 特朗普上周五签署了"大而美"法案，涵盖减税和大规模支出举措，包括延长2017年通过的企业和个人减 税措施，降低企业税，增加军费开支，削减社会保障项目等，计划在未来十年内减税4万亿美元，削减 1.5万亿美元支出。法案还为芯片制造商、能源企业、航空公司、地产开发商和国防军工企业等带来显 著税收优惠或资金支持，但同时削减了电动汽车补贴、可再生能源项目税收抵免，对AI企业监管条款 被删除，并对部分高校和食品企业造成不利影响。美国国会预算办公室估算，该法案未来十年将使美国 赤字增加约3.4万亿美元，减少税收收入，引发财政赤字和债务增加的担忧。 评论员许戈："大而美"税改法案正式通过，特朗普称，这将对美国未来经济与股市产生积极影响。从法 案内容看，未来十年，美国可能进入超级宽松财政刺激，白宫预算委员会预测未来赤字将增加3.4万亿 美元，赤字率也将超过6%，对经济和股市流动性充足可能是利好，但对美债和美元是利空，美债无限 扩大后，每年的利息支出将成为政府巨大负担，对美元汇率可能会是打击，但大宗商品、美股和贵金属 或迎来利好。 评论员徐广语：1）美国"大而美"法案通过将显著冲击 ...

EBITDA 或迎来兑现，重视被低估的 AIDC20250706 摘要 国内 AIDC 上市公司 EV/EBITDA 估值普遍低于国际水平，通常在 10 到 15 倍之间，相较于国际龙头公司 20 到 30 倍的估值存在明显低估，表 明国内市场存在价值重估空间。 成熟 IDC 项目 REITS 发行显示出高估值，润泽科技和万国数据的 REITS 项目 EV/EBITDA 分别约为 14 倍和 17 倍，认购倍数均超 160 倍，与国 内上市公司估值倒挂，突显市场对 AIDC 资产的认可。 国内 AIDC 上市公司 EBITDA 具备持续增长潜力，受益于 2024 年超大 规模 AIDC 招标项目在 2025 年下半年陆续交付，预计 2026 年将迎来 显著增长，且未来有望加速。 国产 ASIC 芯片的推出将增加对 AI DC 资源的需求，因其功耗高于先进 制程芯片，长期来看，将推动 AIDC 行业 EBITDA 持续增长，进一步支 撑国内上市公司估值提升。 可通过短期、中期、长期 EBITDA 预测计算 AIDC 公司市值空间，以 20 倍 EV/EBITDA 为基准，根据公司 EBITDA 规划推算潜在市值 ...

本轮海外 AI 算力大涨的五个关键词 20250706 摘要 北美四大云厂商一季度 AI 算力资本支出同比增长 62%，总额达 719 亿 美元，Meta 上调全年资本支出预期，缓解了市场对 2025 年资本支出 收缩的担忧，表明科技巨头看好 AI 产业趋势。 微软 AI 基础设施 TOKEN 处理量同比增长五倍，谷歌 TOKEN 月均调用 量同比增长 50 倍，字节跳动豆包大模型 TOKEN 日均调用量同比增长 137 倍，显示 AI 应用端对算力需求旺盛。 博通 Q2 单季度 AI 相关业务收入同比增长 46%，Marvell 上调 ASIC 市 场规模指引 37%，反映市场对定制化加速器芯片需求强劲。 英伟达在 AI 大模型训练领域占据重要地位，其 GPU 芯片是底层算力核 心。随着 AI 模型转向推理阶段，英伟达推出 Nvlink 方案并布局低成本 芯片，积极拓展推理侧市场。 AI 大模型推理侧需求快速增长，推动 IC 芯片市场竞争格局变化。除传统 芯片厂商外，谷歌、亚马逊和 Meta 等也纷纷自研或合作部署 ASIC 芯片。 黄仁勋在 VDIO 股东大会上强调 AI 需求强劲，英伟达股价创新高，市值 ...

Investment Rating - The industry investment rating is "Outperform the Market - A" [9] Core Insights - The overseas AI computing sector has experienced a significant rebound, driven by five key factors reflecting optimism in the industry, increasing downstream application demand, supply from upstream chip manufacturers, confidence from major players, and infrastructure investment [7][22] - Major cloud providers in North America reported a total capital expenditure (Capex) of $71.9 billion in Q1 2025, a 62% year-on-year increase, indicating a strong consensus on AI as a key industry trend [3][18] - The rapid growth in token usage, with Google's monthly average token calls projected to increase from 9.7 trillion in April 2024 to 480 trillion in April 2025, signifies a surge in demand for computing power, particularly in AI applications [4][19] - Broadcom, a leader in ASIC chips, reported quarterly revenue exceeding $15 billion, with AI business revenue surpassing $4.4 billion, reflecting a 46% year-on-year growth, as demand for low-cost ASIC chips rises [5][20] - NVIDIA's CEO emphasized the strong and sustained demand for AI, marking the beginning of a decade-long infrastructure buildout, with the company's stock reaching a historic high of $160.98 and a market cap exceeding $3.92 trillion [6][21] Summary by Sections Capex - North American cloud giants reported a total Capex of $71.9 billion in Q1 2025, up 62% year-on-year, with optimistic forecasts for continued investment in AI infrastructure [3][18] Token - Microsoft's Azure AI infrastructure processed over 100 trillion tokens in Q1 2025, a fivefold increase year-on-year, while Google's token usage is expected to grow 50 times within a year, indicating a rapid increase in demand for AI computing power [4][19] ASIC - Broadcom's ASIC chip revenue reached over $4.4 billion, a 46% increase year-on-year, as the demand for AI applications shifts towards more cost-effective ASIC solutions [5][20] NVIDIA - NVIDIA's stock hit a record high, with the CEO highlighting the ongoing strong demand for AI and the company's strategic roadmap for product updates, reinforcing its leadership in the AI computing sector [6][21] Stargate - OpenAI and Oracle announced a significant expansion of their Stargate AI computing infrastructure project, with an initial investment of $100 billion, aiming to enhance AI infrastructure in the U.S. [7][22]

弘则电子 Q2 策略：国产高端化全面出击 20250627 摘要 消费电子需求疲软导致多家公司二季度展望低于预期，尤其手机和 PC 市场，引发对电子板块持续增长的担忧，投资者风险偏好回暖时可能提 前交易长期逻辑。 中美半导体产业存在差异，美股聚焦 AI，业绩兑现度较高，而中国侧重 消费电子，AI 应用效果待观察。2023 下半年至 2024 年，美股资本开 支浪潮与中国消费电子复苏推动半导体周期，但目前可能处于周期下行 阶段。 消费电子一季度增长停滞，AI 资本开支增速放缓，预计下半年 AI 增速将 减缓。工业和汽车半导体市场虽有改善，但对整体半导体市场拉动作用 有限，当前处于周期景气度高位，后续将逐步降速。 互联网公司资本开支接近经营现金流，预示 2026 年资本开支增速可能 受限。定制化 ASIC 芯片放量，将推动小型算力公司增长，定制化推理 芯片份额提升明显。 国内晶圆厂出货量增长，但消费电子需求疲软和产能过剩导致利润率低。 国内半导体股票估值提前反映 AI 预期，但实际进展不如预期，近期估值 回调明显，国产化突破潜力仍在。 Q&A 过去几个月电子板块的市场表现如何？有哪些主要影响因素？ 工业和汽车半导 ...

国金证券近日发布电子行业周报：马斯克发布推特表示，他正在和xAI团队整夜在打磨 Grok，进展不错，下一代模型将直接跳到Grok4，计划在7月4日之后发布，马斯克明确提 到，还需要为专门的编码模型进行一次大规模训练，意味着Grok4或将在代码生成和理解方 面有重大突破。 以下为研究报告摘要： Grok4即将发布，关注二季度业绩有望超预期方向。马斯克发布推特表示，他正在和xAI 团队整夜在打磨Grok，进展不错，下一代模型将直接跳到Grok4，计划在7月4日之后发布， 马斯克明确提到，还需要为专门的编码模型进行一次大规模训练，意味着Grok4或将在代码 生成和理解方面有重大突破。根据路透社报道，OpenAI最近开始租用谷歌的AI芯片，为其 ChatGPT及其他产品提供算力支持。谷歌、亚马逊、Meta等公司ASIC芯片快速发展，谷歌 及亚马逊的ASIC芯片数量今年大幅增长，明年有望持续，Meta计划在2025年Q4发布由博通 设计的ASIC芯片MTIA T-V1，2026年年中，芯片面积翻倍的MTIA T-V1.5将上市，2027年将 发布性能压倒性进步的MTIA T-V2，Meta的目标是在2025年年底和20 ...

Group 1 - The ChiNext Index rose over 1% on June 24, with the Shanghai Composite Index increasing by 0.22% and the Shenzhen Component Index rising by 0.71% [1] - The Sci-Tech Innovation Board Artificial Intelligence ETF (588930) experienced a morning surge, increasing by 1.48%, with significant gains in constituent stocks such as Cambricon Technologies (up over 5%) and others [1] - The ETF has seen a net inflow of over 20 million yuan in the past two days, indicating strong investor interest [1] Group 2 - According to Dongfang Securities, ongoing policy uncertainty is expected to drive the localization of supply chains, benefiting companies in semiconductor equipment, materials, and EDA sectors [2] - Domestic wafer manufacturers and memory chip producers are likely to increase their procurement of localized products to ensure supply chain security and capacity construction [2] - Industrial growth in the AI sector is highlighted by the emergence of innovative applications such as HarmonyOS intelligent agents and Agentic AI, with significant advancements in chip production expected by 2025 [2]