Group 1 - NVIDIA Corporation continues to lead the AI and gaming sectors, with a significant increase in revenue driven by strong demand for GPUs, reporting a year-over-year growth of 101% in Q2 2023 [1] - Micron Technology Inc is facing challenges due to a downturn in the memory chip market, with a projected revenue decline of 50% year-over-year for the fiscal year 2023 [1] - Palantir Technologies Inc is expanding its government and commercial contracts, showing a revenue increase of 22% year-over-year in Q2 2023, indicating strong demand for its data analytics solutions [1] Group 2 - The semiconductor industry is experiencing volatility, with supply chain disruptions and fluctuating demand impacting companies like Micron and NVIDIA [1] - The AI sector is driving growth for technology companies, with NVIDIA positioned as a key player benefiting from increased investments in AI infrastructure [1] - The overall market sentiment remains cautious, with investors closely monitoring earnings reports and guidance from major tech firms [1]

造芯片的还有高手？ 刚刚推出的一款最新芯片，直接冲上硅谷热榜。 峰值推理速度高达每秒17000个token。 什么概念呢？当前公认最强的Cerebras，速度约为2000 token/s。 速度直接快10倍，同时成本骤减20倍、功耗降低10倍。 这就意味着，LLM真正来到了亚毫秒级的即时响应速度。实机效果be like： 但这块一夜之间刷屏硅谷的芯片，并非出自英伟达、AMD之手，而是一家成立仅两年、团队仅有24人的初创公司——Taalas。 芯片代号HC1，也是公司的首款产品。 不同于所有竞争对手，Taalas选择了迄今为止最极端的技术方案—— 模型不再加载到内存里，而是直接刻在硅片上。换言之，芯片即模型。 结果显而易见，Taalas撬动了芯片算力护城河：H100买不到，试试HC1又何妨？ 速度快10倍，功耗降至10分之一 HC1目前搭载Llama 3.1 8B模型，用户每秒最高可生成17000 token/s，远高于主流GPU/ASIC。 其中，在同一模型下Cerebras接近每秒2000个token，SambaNova约为每秒900个token，Groq约为每秒600个，英伟达Blackwell架构的B2 ...

英伟达(NVDA.US)通过扶持CoreWeave构建的"算力-金融"闭环实在太诱人，逼得谷歌(GOOGL.US)和 AMD(AMD.US)不得不"摸着石头过河"，试图用真金白银砸出一个属于自己的AI芯片生态圈。 2月20日，据媒体援引知情人士消息，谷歌正在探索利用其雄厚的财力来构建更广泛的AI生态系统，核 心策略就是向数据中心合作伙伴提供资金支持，条件是对方必须使用谷歌的TPU芯片。与此同时，华尔 街见闻写道，AMD也在通过为客户贷款提供担保的方式，激进地推广其AI芯片。 这一策略被市场视为对英伟达"CoreWeave模式"的直接复刻——即通过扶持"新兴云（Neocloud）"厂 商，绕开已经被英伟达占领或自研芯片的传统云巨头，建立自己的"嫡系部队"。 效仿英伟达路线：注资"新型云"与矿企 知情人士透露，谷歌正在洽谈向云计算初创公司Fluidstack投资约1亿美元。这笔交易对Fluidstack的估值 约为75亿美元。 这种做法虽然能短期推高销量，但也让芯片厂商在AI需求放缓时面临巨大的风险敞口。一旦市场遇 冷，AMD的资产负债表将直接承受冲击。 为何都要绕路走？ 这不仅是一笔财务投资，更是一次战略绑定。 ...

Core Viewpoint - AI spending continues to grow despite market fluctuations, indicating a bullish trend for companies involved in AI, with expectations of sustained growth in the coming years [1] Group 1: Nvidia - Nvidia remains a leading provider of computing units in the AI sector, with significant improvements in its chip architecture, Rubin, which requires fewer GPUs for training and inference [4] - For fiscal year 2027, Nvidia is projected to grow at a rate of 65%, up from 57% in FY 2026, driven by accelerating demand for its GPUs [6] Group 2: Taiwan Semiconductor Manufacturing (TSMC) - TSMC is the largest chip foundry globally and benefits from the AI buildout, expecting nearly 30% revenue growth in U.S. dollars this year [7] - TSMC is well-positioned to capitalize on increased AI spending, with major hyperscalers planning to spend around $650 billion on capital expenditures [8] Group 3: Broadcom - Broadcom is emerging as a competitor in the AI computing landscape, offering ASICs that provide similar or better computing power at a lower price compared to Nvidia GPUs [9] - Demand for Broadcom's custom AI chips is surging, with management projecting revenue from AI chips to double in the upcoming quarter [11] Group 4: Microsoft - Microsoft is currently trading at a lower valuation, with a forward earnings ratio of 24, presenting a potential buying opportunity despite previous investor hesitance [12][14] Group 5: Alphabet - Alphabet has regained its position as a leader in generative AI, with its Gemini model and Google Cloud platform showing impressive growth [15] - While Alphabet may not have the same upside potential as other stocks, it is considered a solid foundation for a portfolio due to its strong prospects in AI innovations [16]

Group 1 - The tech sector is undergoing a transition to AI, with Nvidia positioned to benefit despite recent stock price declines [1][4] - Nvidia's CEO, Jensen Huang, believes the shift to AI will take years, indicating that the sector has not yet reached its peak [5] - Nvidia forecasts fiscal Q4 sales of $65 billion, a significant increase from the previous year's $39.3 billion, suggesting strong ongoing demand for AI [6] Group 2 - Nvidia's market cap is $4.6 trillion, with a current stock price of $189.67 and a gross margin of 70.05% [8] - The company is forming strategic partnerships, including a $5 billion investment in Intel, enhancing its manufacturing capabilities for future semiconductor chips [8] - The stock's forward earnings multiple has dropped to levels not seen since the previous administration's tariff policies, indicating an attractive valuation for potential investors [10]

英国媒体此前报道称，英伟达对OpenAI的300亿美元投资将取代双方去年9月宣布的、为期多年的1000 亿美元合作框架。该框架以"意向书"形式公布，原计划随着OpenAI未来数年算力需求增长，由英伟达 分十次、每次投资100亿美元，以换取重要股权；OpenAI则计划采购数百万枚英伟达人工智能处理器， 支持最高10吉瓦的新算力部署。 据报道，上述合作框架未推进为正式协议。今年1月，有报道称该交易已被"搁置"。新的安排为更直接 的股权投资结构，即英伟达最多投资300亿美元换取OpenAI股份。 2026.02.21 本文字数：967，阅读时长大约1分钟 作者 |第一财经 陈玺宇 美国人工智能巨头OpenAI正向投资者更新长期资本支出规划，计划到2030年实现总计约6000亿美元的 算力支出。该公司同时正推进一轮规模可能超过1000亿美元的融资计划。 据报道，OpenAI此次给出的支出目标数字更低、时间表更明确。公司预计到2030年总营收将超过2800 亿美元，消费者业务与企业业务的贡献大致相当。知情人士表示，目前向投资者展示的支出规划将更直 接地与预期收入增长挂钩。 融资方面，知情人士透露，本轮融资规模可能超过1 ...

来源：中国基金报 万亿外资巨头又调仓了！ 根据美国证券交易委员会（SEC）披露，近日资管巨头贝莱德递交了截至2025年12月31日的第四季度持 仓报告（13F）。 根据13F，贝莱德第四季度持仓总市值为5.92万亿美元，环比增长3.67%。去年第四季度，贝莱德投资组 合中新增了247只个股，清仓165只个股。其中，前十大重仓股占总市值30.41%。 具体来看，前五大买入标的为谷歌-A、谷歌-C、礼来、美光（MU.US）、苹果。前五大卖出标的为标 普500ETF-SPDR CALL（SPY.US.CALL）、ServiceNow（NOW.US）、Strategy（MSTR.US）、AT&T （T.US）、Kellanova （K，现已退市）。 "前十大"科技含量较高 增持英伟达、苹果、微软等 根据13F，贝莱德持有的前十大重仓股科技含量较高，具体包括：英伟达（NVDA.US）、苹果 （AAPL.US）、微软（MSFT.US）、亚马逊（AMZN.US）、谷歌-A（GOOGL.US）、博通 （AVGO.US）、谷歌-C（GOOG.US）、META（META.US）、特斯拉（TSLA.US）、礼来 （LLY.US）等 ...

Group 1 - Recent sentiment towards the Magnificent 7 and software stocks has been negative, leading to significant underperformance in these sectors, primarily due to concerns related to artificial intelligence developments [1][2] - The Magnificent 7 companies, including Amazon, Alphabet, and Microsoft, are recognized as AI leaders, but there are market concerns regarding their increasing capital expenditures, with Amazon planning to spend $200 billion in capital expenditures by 2026, up from $132 billion in 2025 and $83 billion in 2024 [2][3] - The anticipated peak in capital expenditures for 2026 may be premature, as management emphasizes the critical nature of these investments, which could lead to ongoing spending beyond initial expectations [3] Group 2 - Microsoft has seen a three-month performance decline of 15.5%, while the Magnificent 7 group as a whole has decreased by 2.7%, contrasting with the Zacks Tech sector's growth of 1.8% and the S&P 500's increase of 3.9% [4] - Total earnings for 427 S&P 500 companies that have reported are up 12.8% year-over-year, with revenues increasing by 8.8%, indicating a positive trend in overall market performance [6][18] - Nvidia is expected to report significant growth, with projected Q4 earnings of $1.52 per share on revenues of $65.56 billion, reflecting year-over-year growth rates of 70.8% and 66.7%, respectively [10][12] Group 3 - The Magnificent 7 group's earnings are projected to increase by 24.2% in Q4 compared to the previous year, with revenues expected to rise by 18.9%, following a previous growth of 28.3% in earnings and 18.1% in revenues in Q3 2025 [13] - The Magnificent 7 is expected to account for 25.5% of all S&P 500 earnings in 2025, up from 23.2% in 2024 and 18.3% in 2023, highlighting their growing influence in the market [17] - The overall earnings picture for the S&P 500 indicates double-digit earnings growth is expected in 2025 and 2026, suggesting a positive outlook for the market [28]

炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 来源：我是AI人 近日，英伟达13F监管文件中的一系列投资调整，将焦点集中在"增持诺基亚"这一操作上。这家AI算力 巨头不仅清仓安谋控股（ARM）剩余股份、退出Applied Digital等AI生态标的，还大幅增持英特尔、新 建仓新思科技，而其中对诺基亚约2.9%的持股，尤为引人深思。 褪去手机巨头光环的诺基亚，为何能获得黄仁勋的青睐？这场增持背后，藏着黄仁勋怎样的战略考量？ 的核心环节，其构建完整AI基础设施的初心从未改变。 具体来看，黄仁勋的战略布局已形成清晰闭环，增持诺基亚只是其中关键一步：增持英特尔，是为了深 化芯片制造合作，破解AI芯片产能瓶颈，筑牢算力底层根基；增持诺基亚，是为了补齐网络通信短 板，为物理AI、边缘计算等场景提供低时延网络支撑，打通算力落地的"最后一公里"；新建仓新思科 技，则是为了强化芯片设计与算力硬件的协同，降低AI芯片研发成本、提升效率。三者相辅相成，共 同指向"算力-网络-芯片设计"的闭环生态，这正是黄仁勋当前的核心战略方向。 如今的诺基亚早已脱离消费电子赛道，转型为全球领先的电信设备供应商，且正处 ...

Group 1 - OpenAI plans to invest approximately $600 billion in computing power by 2030, with a significant portion allocated for hardware procurement, including chips from Nvidia [1] - The company is pursuing a financing round that may exceed $100 billion, with around 90% of the investment coming from strategic investors, including Nvidia, SoftBank, and Amazon [1] - OpenAI's revenue is projected to surpass $280 billion by 2030, with contributions from consumer and enterprise businesses being roughly equal [1] Group 2 - Nvidia's $30 billion investment in OpenAI will replace a previously announced $100 billion collaboration framework, which was intended to be executed over several years [2] - OpenAI's revenue target for 2025 has been raised to $13.1 billion, exceeding the earlier goal of $10 billion, while the expected cash burn for that year is now estimated at $8 billion, down from $9 billion [2] Group 3 - OpenAI, founded in 2015, transitioned from a non-profit research lab to a mainstream market player after launching ChatGPT, which currently has over 900 million weekly active users [3] - The growth of ChatGPT has faced competitive pressure, leading the company to enter a "code red" state to enhance its chatbot capabilities [3] - OpenAI's programming product Codex has over 1.5 million weekly active users, competing directly with Anthropic's Claude Code [3]