🇨🇳 LATEST: DeepSeek confirms V4 models will run entirely on Chinese silicon using Huawei chips amid US export restrictions, per The Information. https://t.co/QzuKGNd2Qa ...

从5亿到200亿美元，一年估值翻37倍，Reflection AI凭什么？Meta开源模型失去榜首后，硅谷急需一个新的开源模型标杆，这家由两位前DeepMind核心人物创办的公司，被视为如今Neolabs的领头羊。Reflection AI能否扛起开源大旗，成为美国版DeepSeek？ ...

1、 美国总统特朗普表示，即使霍尔木兹海峡仍大体关闭，也愿意结束对伊朗的军事行动。 他认为对伊 朗战争很可能很快结束，其他国家无需美国军事援助即可重新开放霍尔木兹海峡。美国国防部长赫格塞 思表示，美国当前"首要任务"是寻求一项协议，以结束与伊朗的战事。伊朗总统佩泽希齐扬表示，伊朗 愿意结束战争，但前提是其诉求得到满足，尤其是获得不再遭受侵略的保证。 2、伊朗外交部长阿拉格齐表示，当前局势"并非谈判"，而是通过直接渠道或"地区朋友"进行的信息交 换。 伊朗仍收到来自美方代表威特科夫的信息，但这并不意味着谈判已启动，相关交流主要为"警告或 交换观点"。 他指出， 目前伊朗未与任何特定方展开谈判。 阿拉格齐强调，伊朗未对美国提出的15项 提议作出任何回应，也未提出任何提议或条件。 3、 伊朗伊斯兰革命卫队发布公告称，将中东地区与18家美国信息通信技术和人工智能企业有关的公司 机构作为"合法打击目标"，包括英伟达、苹果、特斯拉、谷歌、Meta、微软、思科、惠普、英特尔、 甲骨文、IBM、摩根大通及波音等。 4、 外交部长王毅同巴基斯坦副总理兼外长达尔在北京举行会谈，双方就海湾和中东地区局势交换意见 并提出五点倡议： ...

交易咨询资格号： 证监许可[2012]112 晨会纪要 2026 年 4 月 1 日 联系人：王竣冬 期货从业资格：F3024685 交易咨询从业证书号：Z0013759 研究咨询电话： 0531-68808794 客服电话： 400-618-6767 公司网址： www.ztqh.com [Table_QuotePic] 中泰微投研小程序 请务必阅读正文之后的免责声明部分 [Table_Finance] [Table_Report] 中泰期货公众号 | 2020/ 4/ 1 | | 卒」卒 中 中 小 | | | | --- | --- | --- | --- | --- | | 趋势空头 | 農荡偏空 | 農あ | 農荡偏多 | 趋势多头 | | | 多晶硅 | 工业硅 | 稼校 | | | | 生猪 | 锌 | 20号胶 | | | | 硅铁 | 铝 | 车费 | | | | 红枣 | 棉纱 | 中证500股指期货 | | | | 锰硅 | 燃油 | 沪深300股指期货 | | | | 铁矿石 | 棉花 | 中证1000指数期货 | | | | 焦炭 | 沥青 | 上证50股指期货 | | | | 焦煤 ...

Group 1 - Anthropic experienced a significant code leak involving over 510,000 lines of TypeScript code and several core features of its product Claude Code, raising concerns about its security maturity [2] - OpenAI completed a record $122 billion financing round, achieving a valuation of $852 billion, and plans to go public by the end of the year, with significant investments from Amazon, Nvidia, and SoftBank [2] - Huawei aims for global sales revenue of 880.9 billion yuan and a net profit of 68 billion yuan by 2025, with notable growth in its smart automotive solutions business [3] Group 2 - DeepSeek's services experienced disruptions, possibly related to the upcoming release of DeepSeek V4 [5] - Miniso reported a revenue of 6.254 billion yuan for Q4 2025, a year-on-year increase of 32.7%, and an annual revenue of 21.4438 billion yuan, reflecting a 26.2% growth [6] - Vanke achieved a revenue of 233.4 billion yuan in 2025, delivering 117,000 homes and achieving sales of 134.06 billion yuan [6] Group 3 - Nvidia invested $2 billion in Mellanox Technologies to enhance its AI chip supply chain [9] - Amazon's AWS plans to invest 7 trillion won (approximately $4.6 billion) in South Korea by 2031 to expand its AI and cloud infrastructure [9] - Biogen announced a $5.6 billion acquisition of Apellis Pharmaceuticals to strengthen its position in immunology and rare diseases [10]

Core Viewpoint - DeepSeek has experienced significant service disruptions over three consecutive days, raising concerns about its infrastructure and upcoming model releases [1][5][6]. Group 1: Service Disruptions - DeepSeek's services faced interruptions from March 29 to March 31, with outages lasting approximately 1 hour 48 minutes, 10 hours 13 minutes, and 1 hour 3 minutes respectively [1]. - The service disruptions are speculated to be related to the ongoing gray testing of the upcoming V4 model, as the company prepares for its official release [5][6]. - After resolving the service issues, the API documentation did not include any references to the V4 model, indicating potential challenges in the transition to a higher performance architecture [7]. Group 2: Model Development and Innovations - DeepSeek-OCR 2 was released on January 27, 2026, featuring the innovative DeepEncoder V2 method, which allows AI to process images in a more human-like manner by dynamically rearranging image segments based on logical sequences [8]. - The new model achieved a score of 91.09% on the OmniDocBench v1.5 benchmark, representing a 3.73% improvement over its predecessor, while maintaining low computational costs with visual token limits between 256 and 1120 [8]. - DeepSeek-OCR 2's architecture demonstrates the potential of using language model frameworks for visual encoding, paving the way for a unified multimodal encoder capable of extracting features from images, audio, and text [10].

从2026年2月至今，DeepSeek走过了多个预期发布窗口，仍 未正式发布V4，就在市场期待这一模型新品空降时，等来的 却是DeepSeek成立以来最长的一次服务中断。 作者：钱玉娟 封图：图虫创意 3月31日下午17时，"DeepSeek（深度求索）又崩了"再次冲上热搜。约一天前，DeepSeek刚经历 了一次长达12小时的大规模服务中断。 一位模型技术社区负责人表示，从2026年2月至今，DeepSeek走过了多个预期发布窗口，仍未正式 发布V4，就在市场期待这一模型新品空降时，等来的却是DeepSeek成立以来最长的一次服务中 断。 多名国产模型供应商人士推断：此次只有DeepSeek面向C端的产品服务中断，或与模型迭代过程中 进行灰度测试有关。 1月12日，DeepSeek发布了一篇由梁文锋署名的研究论文，提出了Engram条件记忆架构，还探讨 通过可扩展查找实现条件记忆机制，旨在处理Transformer在记忆、训练稳定性和长上下文上的瓶 颈。2月11日，DeepSeek将现有模型上下文窗口从128K扩展至1M Tokens，知识截止时间更新至 2025年5月。 前述技术社区负责人认为，DeepS ...

梦晨 发自 凹非寺 量子位 | 公众号 QbitAI DeepSeek 崩上热搜！ 宕机持续超过 8小时，写论文的、角色扮演的和心情不好找AI吐槽的人也都崩溃了。 But！这不是一次普通的服务中断，反而 被解读为模型升级的前兆 。 就在宕机前，很多用户报告DeepSeek网页版它变了，模型能力大幅提升。 比如经典的用SVG画鹈鹕骑自行车任务，3月29日的DeepSeek，构图和色彩都明显比一周前强。 模型到底有没有真的升级，答案就藏在对话记录里。 首先是身份标识更新，3月29日的版本会稳定的介绍自己为DeepSeek-V3模型。 一周前，同样的问题，它的自我介绍还是模糊的"我是DeepSeek，一款纯文字AI助手……"，对自己的版本号只字不提。 如果说这种变化也可能是偶然，那么知 识截止日期的变化更有说服力 。 在不开启联网搜搜的情况下，现在它知道25年的美国选举结果。但不知道26年2月的大事，有人据此推测 新版本的知识截止日期可能是26年1 月 。 而且DeepSeek这家公司，不预告、不公告，瞧瞧静默升级模型，已经不是第一次了。 DeepSeek真的变了 最后，3月29日DeepSeek 一次性写代码生成 ...

Listen on the go! A daily podcast of Wall Street Breakfast will be available by 8:00 a.m. on Seeking Alpha, iTunes, Spotify.Getty Images Will an isolationist pivot by the U.S. ultimately increase or decrease global defense spending over the next decade?Click here to take the poll and don't forget to share your thoughts in the WSB comments section. Good morning! Here's the latest in trending:Iran war: The world's most populous country warns of 'considerable downside' to growth and Yemen’s Houthis enter the ...

Jonathan Kitchen/DigitalVision via Getty Images Listen below or on the go via Apple Podcasts and Spotify Mistral raises $830M debt to buy chips for AI data center: report. (00:14) DeepSeek experiences longest outage since launch, going offline for seven hours. (01:14) "Stupidly Cheap": Bill Ackman sees 10X potential in Fannie and Freddie amid Middle East conflict. (01:54) This is an abridged transcript. Mistral AI has raised $830M in new ​debt to buy 13,800 Nvidia (NVDA) chips for a ‌data center near ...