图为展会现场 公司创始人、首席科学家黄小忠教授是国内著名的复合材料专家，毕业于华中科技大学、国防科技大学，现为中南大学博导、中南大学碳化硅纤维复合 材料研究所首任所长。多年来，他瞄准国际上最前沿的技术，针对国家急需高性能、大批量、质量轻、低成本复合材料，他持之以恒在复合材料领域进行科 技攻关，取得了多个重大技术突破，他研发的多个新材料广泛地应用于国防和民用产品上。 黄小忠 教授 泽睿拥有雄厚的人才队伍和技术实力，目前公司员工300 余人，其中有博士学位人员23人、高级职称18人，具有海外留学访学经历科研人员8名，有湖 南省创新人才1名、长沙市科技领军人才1名、长沙市高精尖人才1名，同时有工业部门工作经历的科研骨干和管理人员15人。 公司位于长沙市高新区，现有研发和生产基地近300亩；依托碳化硅纤维及其复合材料湖南省工程实验室、湖南省掺杂碳化硅纤维工程技术研究中心两 个省级科研平台，公司实现了碳化硅纤维产业化突破，建立了规模化生产基地，形成了多系列的碳化硅纤维，现拥有相关专利近40项，具有独立的自主知识 产权，自主可控。 SAMPLE中国2025年会暨第二十届国际先进复合材料制品、原材料、工装及工程应用展览会6 ...

Core Viewpoint - Alibaba's Tongyi Qianwen team has made significant updates to its AI chat product Qwen Chat, enhancing user interaction and adding multiple practical features, aiming to improve product usability and integration within the AI platform [1][4]. Group 1: Product Updates - Users can now start conversations directly from the Tongyi Qianwen homepage, reducing usage costs and enhancing accessibility [3]. - Qwen Chat integrates multiple functionalities, including "in-depth research," "image generation," "web development," "deep thinking," and "search," allowing users to generate high-quality images from text descriptions and assist front-end engineers in coding [3]. - A new desktop client has been introduced, enabling cross-application calls and automated task execution, which enhances work efficiency by bridging different AI models and external tools [3]. Group 2: Competitive Positioning - The Qwen model family has been continuously updated, with the latest Qwen3 model outperforming top global models in various benchmark tests, showcasing its competitive edge [4]. - Alibaba is addressing the "product strength" gap of its large models by enhancing usability and user perception, indicating a strategic intent to build a unified AI platform that is user-friendly and feature-rich [4]. Group 3: Market Presence and Financial Performance - Tongyi Qianwen is recognized as the largest open-source model globally, with Alibaba Cloud holding a 23% market share in China's AI infrastructure (AIIaaS) market, surpassing the combined share of its closest competitors [5]. - The "cloud + AI" strategy has become a new growth engine for Alibaba, with the cloud intelligence group achieving a revenue of 1180.28 billion yuan in the 2025 fiscal year, marking an 11% year-on-year increase [5]. - AI-related product revenues have seen triple-digit growth for seven consecutive quarters, indicating strong demand and market penetration in traditional vertical industries [5]. Group 4: Investment and Future Plans - Alibaba plans to invest 380 billion yuan in AI infrastructure over the next three years, exceeding its total technology investment over the past decade [6]. - The company has announced the issuance of zero-coupon exchangeable bonds to fund cloud computing infrastructure and support international expansion [6].

Core Insights - Alibaba and JD.com represent two distinct strategies in capturing the digital commerce market, with both companies expanding beyond traditional retail into cloud computing, logistics, AI, and global markets [2][3] - Recent strategic moves and financial performances of both companies highlight contrasting adaptation and growth potential in a competitive landscape [2][3] Alibaba's Position - Alibaba's transformation into an AI-powered technology platform marks a significant shift, enhancing its growth prospects [4] - The partnership with HONOR for integrating AI agents into consumer hardware opens new monetization pathways and expands Alibaba's influence [4][5] - A commitment to invest RMB 380 billion over three years in AI infrastructure reflects strong confidence in this strategic direction [5] - The Qwen3 language model's adoption across various platforms validates the commercial viability of Alibaba's AI strategy [5][6] - The cloud division is experiencing accelerating revenue growth due to rising AI demand, with over 300 million global downloads of Qwen [6] - Core e-commerce platforms show resilient user engagement and improved monetization efficiency through AI enhancements [7] - The Zacks Consensus Estimate for fiscal 2026 earnings is projected at $10.14 per share, indicating a 12.54% year-over-year increase [8] JD.com's Challenges - JD.com faces strategic misalignment, particularly in its aggressive food delivery expansion, leading to cash outflows and unsustainable profitability [11] - International logistics investments in Abu Dhabi and the UK are seen as poorly timed, straining resources amid domestic competition [12] - JD's historical logistics advantage is becoming commoditized, with partnerships lacking transformative impact [13] - The launch of JoyExpress in Saudi Arabia exemplifies JD's capital-intensive expansion approach, which may limit scalability [14] - The consensus estimate for 2025 earnings is pegged at $3.6 per share, indicating a 15.49% year-over-year decline [14] Valuation and Performance Comparison - Both companies trade at discounted valuations, with Alibaba's P/E ratio at 10.02x compared to JD's 7.66x, reflecting Alibaba's superior financial metrics and growth prospects [15] - Recent price performance shows Alibaba shares have increased by 25.8% year-to-date, while JD.com has declined by 10%, indicating investor confidence in Alibaba's strategy [18] Conclusion - Alibaba is positioned as the superior investment choice due to its transformative AI capabilities, diversified revenue streams, and strategic growth positioning [20] - JD's capital-intensive growth strategy and competitive challenges present significant headwinds, suggesting a potential reduction in exposure to JD [20]

近来老婆不知道又被谁忽悠了，说是律师行业赚钱特简单，于是买了一大堆书，竟真的学了起来： 想着它看书也可以少烦我，本着多一事不如少一事的心态，便没有去阻止，只不过她果然没坚持一个月就结束了，对此我是极为"惋惜"的！ 事实上，从去年到今年，我已经为10多家企业以及40多位个人提供过知识咨询服务了，并且反馈都是很不错的。 但是，我确实是无力在任何角度说服我老婆的，几乎到了一说就错的地步，但大家可能没想到： 事实上，我老婆这种画像，其实本应该是我重点收割的韭菜，她这种类型的咨询者，去年我拒绝了不下10个！ 大模型时代来临，他带来了信息平权，但信息平权也救不了"焦虑的傻瓜"，因为他们根本无力去分别知识的好坏，比如以下爆款课程： 《财富自由课》、《短视频爆款秘籍》、《Python30天从入门到精通》、《如何靠写作月入十万》、《AI爆文赚钱功率》…… 我可以大胆下结论：学的是无用的垃圾，耗费的是精力和热情，交的是智商税！ 因为这一切我太熟悉了…… 瞎努力 大学时期，在做了几个项目后，技术很快的达到了一个瓶颈期，于是陷入了成长焦虑。 我认为自己要专精底层技术，于是不知道在图书馆哪里摸出来一本《数据库底层结构原理》的书读了一学 ...

Core Insights - TSMC reported June revenue of NT$263.71 billion, a decrease of 17.7% month-over-month but an increase of 26.9% year-over-year [1] - For the first half of the year, TSMC's revenue reached NT$1,773.046 billion, a 40.0% increase compared to the same period last year [1] - Q2 revenue was NT$933.796 billion (approximately US$31.95 billion), up 38.6% year-over-year, exceeding company guidance and market expectations [1] - Analysts had predicted Q2 revenue of NT$927.831 billion, while JPMorgan expected US$29.95 billion, with a quarter-over-quarter growth of 17% [1] - Strong demand for 3nm and 5nm processes, along with urgent orders for older process nodes, contributed to the revenue growth despite significant appreciation of the New Taiwan Dollar [1] - TSMC's gross margin for Q2 is expected to remain at 57.9% [1] Company Outlook - TSMC's CEO, C.C. Wei, indicated that while U.S. tariffs have had some impact, demand for AI remains strong and continues to exceed supply [2] - The company forecasts a mid-teens percentage growth in revenue for 2025 [1][2] - TSMC expects full-year revenue growth in USD to be close to the mid-20% range (24%-26%) [2] - The company plans to hold an earnings call on July 17, focusing on order outlook for the second half, capital expenditure plans for 2026, and progress on 2nm process technology [2] - TSMC benefits from the AI boom as the largest chip foundry, producing high-end processors for companies like NVIDIA and AMD [2] - The company announced plans to invest an additional US$100 billion in the U.S., following a previous commitment of US$65 billion for three factories [2]

LSEG SmartEstimate（来自 21 位分析师的调查）预测，台积电二季度营收为9,278.31 亿新台币。 公司6月销售额2,637.1亿新台币，较去年同期增加了26.9%。根据稍早前公布的财务数据计算，台积电二季度销售额为9337.96亿元新台币，约合 319.5亿美元，远超公司指引和市场预期。 台积电二季度销售额超预期大增，再次佐证了AI需求的强劲。 台积电周四最新公布报告显示，公司6月销售额2,637.1亿元新台币，较上月减少了17.7%，但较去年同期增加了26.9%。报告还显示，1-6月累计销 售额为17730.46亿元新台币，较去年同期增加了40.0%。 | 項目 | 2025年 | 2025年 | 月增(減) | 2024年 | 年增(減) | 2025年 | 2024年 | 年增(減) | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | 6月 | 5月 | % | 6月 | % | 1至6月 | 1至6月 | % | | 營收 | 263,709 | 320,516 | (17.7) | 207,869 | 26 ...

执业证书编号：S0740519080001 Email：wangke03@zts.com.cn | 基本状况 | | | --- | --- | | 总股本(百万股) | 122.24 | | 流通股本(百万股) | 122.24 | | 市价(元) | 100.77 | | 市值(百万元) | 12,317.67 | | 流通市值(百万元) | 12,317.67 | 股价与行业-市场走势对比 自动化设备 1、《工业领域核心标的，两大业务景 气反转》2025-03-06 奥普特（688686.SH） 证券研究报告/公司点评报告 2025 年 07 月 10 日 | 评级： | 买入（维持） | | 公司盈利预测及估值 | | | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | 指标 | | 2023A | 2024A | 2025E | 2026E | 2027E | | 分析师：王可 | | | 营业收入（百万元） | | 944 | 911 | 1,103 | 1,331 | 1,595 | | ...

而YU7的订单量太多，本来就招人羡慕嫉妒恨了，这下找到了黑点了，大家一涌而上，疯狂一顿输 出…… 那么汽车上能不能用消费级芯片？要看用在什么地方，像汽车的控制、驾驶等芯片肯定不能用消费级 的，必须是车规级的。 但智能座舱芯片，用消费级的问题不大，因为智能座舱其实控制的是就是车内那一个"大平板"，并不参 与车辆的控制，比如特斯拉等，也在座舱芯片上使用消费级芯片，所以小米YU7使用骁龙8Gen3也并不 是不可接受。 座舱芯片的使用环境，并不很严苛的，消费级芯片的要求，其实也是基本能够符合的。 事实上，对于小米使用车规级芯片这事，可能很多人并没有看到，背后其实藏着雷军一个大大的野心， 那就是未来用自研来替代。 汽车座舱芯片可不便宜，比如高通的8295P之类的，几千块钱一颗，还性能一般般。 前段时间，网上铺天盖地的吐槽小米YU7使用消费级芯片。意思就是YU7上使用的那颗骁龙8Gen3，是 用于手机上的芯片，而不是车规级芯片。 而小米目前有自研消费级芯片，比如玄戒O1，它就是消费级芯片，如果小米能够将它用于汽车上，对 于小米的成本而言，就只有几百块了。 一台车就节省上千块，甚至几千块的成本，100万台车就是几十亿的成本 ...

以下文章来源于光子星球 ，作者郝鑫 光子星球 . 细微之处，看见未来 将投中网设为"星标⭐"，第一时间收获最新推送 处在上市前夜的AI六小虎，却各有各的难。 作者丨 郝鑫 编辑丨 吴先之 来源丨 光子星球 隔壁阿里Qwen3刷新模型记录，字节豆包全家桶打折甩卖。资本和市场就像两个无形的手，让大模 型六小虎们（月之暗面、智谱AI、MiniMax、阶跃星辰、百川智能、零一万物）不敢再沉寂下去。 智谱宣布获得浦东创投集团和张江集团联合10亿元人民币的战略投资，并开源新一代视觉语言大模 型GLM-4.1V-Thinking和全新的生态平台"Agent应用空间"；月之暗面开源代码模型Kimi-Dev， 上市这个选择，既关乎"面子"也关乎"里子"，一方面是靠融资上市续命，缓解"吞金"的压力，另一 方面是争夺行业话语权，撕掉"掉队"的标签。 不过，处在上市前夜的AI六小虎，却各有各的难，能否一举成功上市尚未可知，上市之后仍是挑战重 重。 各有各的难处 首个AI Agent项目Kimi Researcher（深度研究）开启内测；MiniMax"五连更"，涵盖推理模型、 视频生成模型、智能体、音色设计。 截至目前，六小虎中的两 ...

Polaris团队 投稿 量子位 | 公众号 QbitAI 4B模型的数学推理能力和顶尖商业大模型差在哪里？ 香港大学NLP团队联合字节跳动Seed、复旦大学发布名为 Polaris 的强化学习训练配方： 通过Scaling RL，Polaris让4B模型的数学推理能力（AIME25上取得79.4，AIME24上取得81.2） 超越了一众商业大模型 ，如Seed-1.5- thinking、Claude-4-Opus和o3-mini-high(25/01/31)。 并且，Polaris-4B的轻量化允许在消费级显卡上部署。 详细的blog、训练数据、模型和代码都已 全部开源 ，链接可见文末。 围绕待训练模型进行参数配置 之前的RL训练配方，如DeepScaleR，已经展示了Scaling RL在较弱基模型上强大的效果。 但对于目前最前沿的开源模型（如Qwen3），Scaling RL是否也能复现如此显著的提升呢？ Polaris的研究团队给出了明确回答：可以！ 具体来说，Polaris通过仅仅700步的RL训练，成功地 让Qwen3-4B在数学推理任务上接近了其235B版本的表现 。 只要方法得当，RL还存在 ...