马克斯引用了自己在2000年1月互联网泡沫时期写的一份备忘录，以及2025年1月发布一份题为《警惕泡 沫》的报告，来阐述市场过度乐观或过度悲观时的风险。 "当多数投资者乐观时，他们会推高价格，甚至可能使其超越价值；而当悲观情绪主导时，价格又会跌 破价值。"他写道。 尽管面临特朗普关税政策的不确定性，美股今年仍延续了去年的火爆行情。自年初以来，标普500指数 已经累计上涨了逾10%，而少数几只大型科技股贡献了其中大部分涨幅。 在此背景下，有一些分析师开始质疑美股"科技七巨头"估值过高。 不过，华尔街价值投资大师、橡树资本管理公司的联合创始人霍华德•马克斯并不认同这一观点。他认 为，美股"科技七巨头"估值并没有过高。相反，他认为标普500指数的其它成分股可能存在问题。 马克斯是橡树资本管理公司（Oaktree Capital Management）的联合创始人，该公司以投资不良信贷和高 收益债券而闻名。据公开消息，截至2024年底，橡树资本资产管理规模为2020亿美元。 马克斯著有《投资最重要的事》《周期》等多本在投资领域颇具影响力的书籍，他所展现出的投资智慧 赢得了"股神"沃伦·巴菲特的认可。巴菲特曾大力推荐《 ...

Investment Rating - The investment rating for the industry is "Outperform the Market" (maintained rating) [1] Core Insights - The capital expenditures (CapEx) and performance of major cloud vendors such as Microsoft, Meta, Amazon, and Google have shown significant growth in Q2 2025, indicating an acceleration in downstream cloud demand. Meta and Google have raised their full-year CapEx forecasts, reflecting this trend [2] - The Software as a Service (SaaS) sector has demonstrated strong performance, with a median revenue exceeding expectations by 2.8%, marking the highest level since Q2 2022. The annual recurring revenue (ARR) for reported companies reached $2.187 billion, a 106.1% increase year-over-year [2][71] - Despite strong performance, software stock prices have been affected by concerns over AI disruption, with the BVP Nasdaq Emerging Cloud Index dropping over 8% since July 31 [2][71] Summary by Sections 1. Cloud Vendors' CapEx and Performance Review - In Q2 2025, Microsoft reported a CapEx of $24.2 billion, a year-over-year increase of 27.4% and a quarter-over-quarter increase of 13.1%. Over 50% of this spending was directed towards long-term assets related to cloud computing and AI [9] - Microsoft’s revenue reached $76.441 billion, a year-over-year growth of 18.1%, with Azure cloud revenue growing 39% [15] - Meta's Q2 2025 CapEx was $17 billion, a 100.8% increase year-over-year, with a full-year CapEx forecast raised to between $66 billion and $72 billion [22] - Google reported a CapEx of $22.446 billion, a 70.23% year-over-year increase, with expectations for 2026 CapEx to rise to $85 billion [35] - Amazon's Q2 2025 CapEx reached $31.4 billion, a 91.5% increase year-over-year, primarily for AWS-related investments [46] 2. SaaS Vendors' Performance Review - The SaaS sector's overall revenue median exceeded expectations, with a notable increase in ARR, reflecting a strong upward trend in performance [2][71] - The current EV/NTM revenue median for SaaS companies is 5.1 times, indicating high valuations despite growth slowdowns [76] - The software sector's stock performance has been under pressure due to AI disruption concerns, despite strong earnings reports [71]

亿万富翁Ackman旗下潘兴广场二季度持有超过580万股亚马逊股票，价值约12.8亿美元。这是该基金首次建仓 亚马逊，成为其第四大重仓股。Ackman的入场时机恰逢亚马逊股价在低位，还大幅增持了谷歌母公司 Alphabet股票，增幅达20.84%，凸显出其对AI技术商业化前景的乐观预期。 亚马逊股价今年以来已反弹5.3%，此前曾因市场对生成式AI热潮和美国关税政策的担忧而大跌超过30%。 Ackman的入场时机恰逢股价低位 ，体现了其逆向投资的一贯风格。 亿万富翁、对冲基金大佬Bill Ackman二季度大科技股，新建价值近13亿美元的亚马逊头寸，大幅增持谷歌母公 司Alphabet，凸显其对AI领域投资机会的看好。 8月14日，据提交给美国SEC的监管文件13F显示，Bill Ackman旗下潘兴广场资本管理公司（Pershing Square） 在二季度末持有超过580万股亚马逊股票，价值约12.8亿美元，占其投资组合的9.31% 。这是潘兴广 场首次建仓亚马逊，成为其第四大重仓股。 据监管文件显示， 截至6月底，Pershing Square持有582.3316万股亚马逊股票，总价值约12.8亿美元，占 ...

华尔街最大的几家对冲基金——桥水基金、老虎全球和Discovery Capital——在第二季度加大了对大型 科技公司的投资力度，因为人工智能领域的增长呈现出前所未有的态势。 科技股繁荣 这与今年早些时候的情况形成了鲜明对比。当时，由于贸易关税导致金融市场波动，顶尖基金经理对大 型科技公司的投资预期变得悲观。投资者对通胀上升的担忧以及对人工智能泡沫引发"七大巨头"股票抛 售的恐惧，使得他们纷纷抛售相关股票。 但自那以后，科技股实现了大幅反弹。今年截至目前，得益于大型科技股的涨势，标准普尔500指数已 上涨10%，而这些公司的市值占该指数所涵盖公司总市值的近三分之一。 以下是顶级对冲基金持股情况的变化详情： 桥水基金 在第二季度，桥水基金增持了英伟达、Alphabet以及微软的股票。 由达里奥创立的宏观对冲基金在英伟达股票上的投资增加了两倍多。截至6月底，该基金持有英伟达723 万股股票，较3月底增加了154.5%。英伟达还是桥水基金在单一股票上的最大投资，总投入金额达11.4 亿美元。 此外，该基金在Alphabet和微软的持股分别上涨了84.1%和111.9%，总计达到9.87亿美元和8.53亿美元。 其他 ...

智通财经APP获悉，投资大佬段永平美股持仓主体H&H International Investment披露了截至2025年6月30日的第二季度持仓报告(13F)。据统计，H&H二季度 美股持仓总市值达115亿美元，上一季度总市值为120亿美元，环比下降4%。H&H的持仓组合中新增了0只个股，增持了4只个股，减持了4只个股，清仓了1 只个股。其中前十大持仓标的占总市值的100%。 | 2025-06-30 | | | --- | --- | | 13F Activity | | | Market Value | $11.5b, Prior: $12.0b | | Inflows (Outflows) as % of Total MV | +3.6782% | | New Purchases | 0 stocks | | Added To | 4 stocks | | Sold out of | 1 stocks | | Reduced holdings in | 4 stocks | | Top 10 Holdings % | 100% | 阿里巴巴(BABA.US)位列第五，持仓约375万股，持仓市值约4.2 ...

时令 发自 凹非寺 量子位 | 公众号 QbitAI 谷歌开源Gemma 3 270M闪亮登场！ 只需几分钟即可完成微调，指令遵循和文本结构化能力更是惊艳，性能超越Qwen 2.5同级模型。 此模型小巧又高效，可以直接在浏览器里本地运行，不用联网，也能生成有创意的内容，比如睡前故事。 发布当天，网友也懵了：以为是270B，结果居然才0.27B。 不仅如此，还有人使用这款迷你模型构建了自己的OCR应用程序。上传一张图片或PDF文件，即可用LLM即时将其转换为结构化的 Markdown格式。 值得一提的是，新模型只有 4个注意力头 ，比Qwen 3 0.6B少12个，真是切实符合其轻量化的定位。 下面让我们一起看看这款迷你Gemma 3到底有哪些亮点？ Gemma 3 270M核心功能 Gemma 3 270M充分体现了这种" 为工作选择合适工具 " 的理念。 作为一款基础模型，它开箱即可精准遵循指令，而微调能彻底释放其真正实力。 经过专门优化，它在文本分类、数据提取等任务中，都能做到准确、快速且成本可控。 简单总结，新模型的核心功能可概括为以下4部分： 紧凑且高效的架构 这款新模型共包含2.7亿参数，其中1.7 ...

Core Viewpoint - Google has officially released the latest model of the Gemma 3 series, named Gemma 3 270M, which is a compact language model designed for specific task fine-tuning, featuring strong instruction tracking and text structuring capabilities [2][3]. Model Features - Gemma 3 270M has 270 million parameters, with 170 million embedding parameters and 100 million in the Transformer module, allowing it to handle specific and rare tokens effectively [7]. - The model is highly energy-efficient, consuming only 0.75% of battery power during 25 dialogues on the Pixel 9 Pro mobile SoC, making it the most energy-efficient model in the Gemma series [7]. - It includes a pre-trained instruction-following model that can be used out-of-the-box for general instructions, although it is not designed for complex dialogue use cases [7]. - Quantization-aware training (QAT) checkpoints are available, enabling the model to run at INT4 precision while minimizing performance degradation, which is crucial for deployment on resource-constrained devices [7]. Practical Applications - Gemma 3 270M is suitable for high-capacity, well-defined tasks such as sentiment analysis, entity extraction, query routing, unstructured to structured text processing, creative writing, and compliance checks [12]. - It can significantly reduce inference costs and provide faster user responses, making it ideal for tasks with high latency requirements [12]. - The model's compact size allows for rapid fine-tuning experiments, enabling users to find optimal configurations in hours rather than days [12]. - It can run entirely on-device, allowing for the development of applications that handle sensitive information without sending data to the cloud [12]. - The model is also designed for creating and deploying multiple custom models, each trained for different tasks without exceeding budget constraints [12]. Market Impact - Google emphasizes that Gemma 3 270M serves as a high-quality foundational model that can be utilized for specialized tasks, leading to efficient production systems [11]. - The model has already shown success in real-world applications, such as the collaboration between Adaptive ML and SK Telecom, where a fine-tuned Gemma 3 4B model outperformed larger proprietary models in specific tasks [11]. - As of last week, the cumulative download count for the Gemma series has surpassed 200 million, indicating strong market interest and adoption [14].

Core Viewpoint - H&H International Investment, managed by Duan Yongping, has made significant adjustments to its portfolio, notably increasing holdings in AI-related stocks like Google and NVIDIA, while also adding to positions in Apple and Pinduoduo, indicating a strong belief in AI's future potential and a focus on high-conviction investments [1][3][8]. Group 1: Portfolio Adjustments - H&H increased its stake in Google by 830,000 shares, a 75.17% increase, and in NVIDIA by 320,000 shares, a 49.56% increase, reflecting a strong commitment to AI investments [1][8]. - The firm also added 894,426 shares of Apple, marking a 2.61% increase, reversing a trend of previous reductions, which underscores a renewed confidence in Apple's long-term value [5][9]. - Pinduoduo saw an increase of 908,794 shares, a rise of 11.72%, solidifying its position as the third-largest holding in the portfolio [9]. Group 2: Sector Focus - The adjustments indicate a concentrated investment strategy in AI, with a clear preference for leading companies in the sector, while showing a selective approach by reducing positions in other tech giants like Microsoft and TSMC [8][9]. - The portfolio's total value reached $11.53 billion, with Apple representing over 62% of the total holdings, followed by Berkshire Hathaway at approximately 14.2% [3][6]. Group 3: Market Sentiment - The moves reflect a broader market sentiment towards AI, with Duan Yongping's actions signaling a strong belief in the growth potential of AI technologies, while also demonstrating a cautious approach by not fully committing to all major tech players [8][10].

Core Viewpoint - The article emphasizes that while GPT-5 is receiving significant attention, the true focus should be on Google DeepMind's Genie 3, which represents a breakthrough in world modeling technology that could reshape the AI landscape [2][5]. Summary by Sections Introduction - The AI community is currently focused on GPT-5, but there is a risk of overlooking Genie 3, which is considered more significant [2]. World Model Definition - World models generate interactive and logically consistent environments, allowing users to explore and interact, unlike traditional video which is static and fixed [6]. Genie 3 Demonstration - Genie 3 can create a persistent world where changes made by users are retained, showcasing its ability to maintain logical consistency [9][11]. Disruptive Potential of World Models - World models could democratize high-quality content creation, significantly reducing costs in gaming and film production, and have potential applications in robot training [14][20]. Applications in Autonomous Driving - World models can generate training scenarios for autonomous vehicles, allowing for efficient data generation that adheres to physical laws, thus lowering training costs [15][19]. Relation to Metaverse and Mirror World - The advent of world models could lower the production costs associated with the metaverse, making it more feasible and aligning with the concept of mirror worlds that blend reality and virtuality [20]. Future Investment Opportunities - Companies and investors interested in autonomous driving, robotics, and immersive virtual experiences should closely monitor developments in world modeling technology, as it is seen as a key driver for these industries [22].

Core Viewpoint - The semiconductor industry in China showed a modest increase in July 2025, with a 3.06% rise, lagging behind the Shanghai Composite Index's 3.54% increase, indicating a relatively weak performance in the sector [1][2]. Group 1: Semiconductor Industry Performance - In July 2025, the domestic semiconductor industry (CITIC) rose by 3.06%, while the Shanghai Composite Index increased by 3.54% [1][2]. - Integrated circuits increased by 2.74%, discrete devices rose by 5.05%, semiconductor materials fell by 0.11%, and semiconductor equipment saw a 4.87% increase [1][2]. - Year-to-date, the semiconductor industry (CITIC) has increased by 11.96% [1][2]. - The Philadelphia Semiconductor Index rose by 1.11% in July 2025, while the Nasdaq 100 increased by 2.38%, with the Philadelphia Semiconductor Index up 12.6% year-to-date [1][2]. Group 2: Global Semiconductor Sales and Demand - Global semiconductor sales in June 2025 grew by 19.6% year-on-year, marking the 20th consecutive month of growth, with a month-on-month increase of 1.5% [3]. - The demand for consumer electronics is gradually recovering, while AI hardware infrastructure demand remains strong [3]. - Global smartphone shipments in Q2 2025 remained flat year-on-year, with a 4% decline in shipments from mainland China [3]. - The capital expenditure of the four major North American cloud providers (Google, Microsoft, Meta, Amazon) increased by 69% year-on-year and 23% quarter-on-quarter in Q2 2025, totaling $87.4 billion [3][5]. Group 3: Investment Opportunities - The U.S. has implemented stricter export controls on AI computing chips, while NVIDIA has received approval to sell H20 chips to the Chinese market [4]. - There is a significant demand for domestically produced AI computing chips in China, indicating a broad market space for domestic AI chip manufacturers [5]. - The four major North American cloud providers are expected to continue increasing their capital expenditures, with Google raising its 2025 budget from $75 billion to $85 billion, and Meta increasing its budget to $66-72 billion [5].