搜索“NVIDIA”

Vellante 创立，已打造出一个充满活力的行业领先数字媒体品牌生态，触达 1500 多万精英科技专业人士。我们全新的专有 theCUBE AI Video Cloud 正在受众互动领域开辟新天地，利用 theCUBEai.com 神经网络帮助科技公司做出数据驱动的决策，并始终站在行业对话的前沿。 最新报道 Google 推出 TorchTPU，力图打破 Nvidia 在 AI 算力领域的垄断 SpaceX 股价飙升，Elon Musk 公布碾压预期的盈利和营收 Elon Musk 在战火中的乌克兰限制了 Starlink 服务 Arda Industries 融资 2.32 亿美元，让生产线更具适应性 Eka Robotics 以 14.5 亿美元估值融资 1.5 亿美元，打造可感知力的机械手 OpenAI 收购 Orchestra Bio，将 AI 智能体引入药物研发 最新报道 Google 推出 TorchTPU，力图打破 Nvidia 在 AI 算力领域的垄断 AI - 作者 MIKE WHEATLEY . 2 小时前 SpaceX 股价飙升，Elon Musk 公布碾压预期

匹配字段正文

gone from asking the White House in March 2025 to keep China under strict export controls and widen the list of restricted chips to staying silent as Congress prepares new limits. Nvidia, which is lobbying against the restrictions, is also OpenAI’s chip supplier, a $30B equity investor, and the guarantor of up to $105B for the first phase of an Ohio data center OpenAI will lease. Anthropic has taken the opposite position, backing several of the chip bills and leaving a major tech trade associati

匹配字段正文

ted with technical assistance and editorially reviewed before publication. . „Drei Millionen Grafikkarten, bitte”, sagen Alibaba, Bytedance und Tencent laut Reuters, „Und zwar die Nvidia RTX PRO 5500.” Macht zum Listenpreis rund 35 Milliarden US-Dollar. Weitere chinesische Unternehmen haben demnach sogar noch mehr bestellt, insgesamt also mehr als sechs Millionen. Die Grafikkarten sollen nicht für schöne Spielerlebnisse sondern für KI-Inferenz eingesetzt werden. Allerdings werden die potenzielle

匹配字段正文

hropic的关系，这番话被解读为对竞争对手的暗讽。 安全很重要，但不能以落后为代价 据报道，Altman在内部抱怨称，针对AI发展“限速”的可行提案太少。与此同时，OpenAI眼下承受不起在竞争中落后的代价。据报道，OpenAI和Anthropic都在推进首次公开募股，OpenAI计划在明年之内决定走哪条路。 内部压力也在积聚。来自OpenAI、Anthropic、谷歌和Meta的模型违背自身安全设置，侵入了外部和内部系统，有时出现严重失准。此后OpenAI一直在努力制定新的防护措施，而在三名安全研究员被解雇后，一些员工对公司的安全策略提出了质疑。 在政策层面，美国政府迄今依赖对新模型的自愿预审。6月，美国以担心Anthropic的Fable模型可能存在越狱漏洞为由对其实施出口管制，Anthropic随后暂时关闭了访问。 与此同时，OpenAI、谷歌和Anthropic正在组建前沿AI标准局，但据报道，它们在一些关键问题上仍无法达成一致。Meta CEO Mark Zuckerberg和Nvidia CEO Jensen Huang对此表示反对，认为这类规则主要会让现有玩家免受竞争。

匹配字段正文

titutionen. Ein Bann für verschiedene Einrichtungen wurde immer wieder diskutiert und teilweise auch schon umgesetzt. China hingegen hat erst vor wenigen Tagen einen Entwurf veröffentlicht, demzufolge der Export von bestimmten KI-Technologien untersagt werden soll. Unterdessen arbeitet DeepSeek daran, Chinas Abhängigkeit von Nvidias Chip-Ökosystem zu verringern – mit Programmierwerkzeugen für Huaweis KI-Beschleuniger der Ascend-Reihe, die das Unternehmen als Open Source veröffentlicht hat. (rie)

匹配字段正文

inese AI lab has faced scrutiny from governments and regulators across the globe since its R1 reasoning model shook the AI industry in January 2025, triggering a sharp sell-off in Nvidia shares. The U.S. accused DeepSeek and five other Chinese AI companies, including Alibaba and Moonshot AI, of using large-scale model distillation to extract information from American AI systems, according to a September report. In addition, OpenAI and Anthropic have separately raised concerns about DeepSeek’s al

匹配字段正文

to have the technology removed. Over the summer, the US government considered steps against Chinese open-weight models , according to the New York Times. The options included sanctions, a trade blacklist, and rules on which models federal agencies are allowed to use. The push came after a wave of powerful Chinese models such as Kimi K3 from Moonshot AI. The government pulled back for now after Nvidia, Microsoft, Google, and Meta objected, while OpenAI and Anthropic had lobbied for restrictions.

匹配字段正文

路透社：英伟达已告知客户和合作伙伴，由于中国科技企业的需求远超计划产量，公司无法保证能够满足这些订单图像来源: Justin Sullivan/Getty Images/AFP （德国之声中文网）据路透社周二（10月20日）援引两名知情人士报道，中国科技巨头字节跳动、阿里巴巴和腾讯已向美国芯片制造商英伟达（Nvidia）订购了超过300万块新款RTX PRO 5500显卡，按标价计算总额超过350亿美元（约合310多亿欧元）。这款芯片性能不及英伟达最先进的产品，但符合美国对华出口管制规定。 报道称，英伟达已告知客户和合作伙伴，由于中国科技企业的需求远超计划产量，公司无法保证能够满足这些订单。上述消息人士因未获授权对媒体发言而要求匿名。 需求远超产能 或仅能交付一成 据路透社报道，一名消息人士透露，英伟达向合作伙伴表示，他们最终可能只能拿到所订数量的5%至10%。另一名消息人士称，英伟达表示2027年全球范围内的RTX PRO 5500产量最多只有300万块。两名消息人士说，来自中国客户的订单总量已超过600万块。 字节跳动、阿里巴巴和腾讯三家公司所订显卡的单价约为1.14万美元（约合1

匹配字段正文

且开始为平台能力支付里程碑，为长期研发合作锁定优先权。 中信建投最新研报指出，AI4S正由模型能力竞赛迈向产业化临界点，技术、产业、需求、政策四重共振，价值锚点由模型能力向科研生产力、数据资产与系统化生物设计升维。AI制药产业链已演化为以平台和技术中台为核心、上中下游深度协同的网状赋能体系。 礼来（Eli Lilly）的AI投资版图正是这一趋势的最佳注脚。从NVIDIA的算力底座、到Insilico的小分子生成、到Chai的生物药设计、到Profluent的基因编辑酶、再到Kelonia的体内基因递送——礼来正在构建一个AI-native制药闭环。这一次剂泰科技为礼来补上了这个闭环的最后一环：递送。 但这一事件的投资意义，远不止于一单BD。礼来买的不只是一个LNP配方，而是剂泰NanoForge平台作为Biological AI基础设施的系统能力。这意味着，剂泰科技的估值逻辑，正从管线公司向平台基础设施公司迁移——而港股市场，尚未对此充分定价。 01 算力层，礼来与英伟达共建LillyPod和AI Co-Innovation Lab，5年内共同投资最高10亿美元。生物药设计层，投资Ch

匹配字段正文

t weeks after launching Amazon introduces toolkit for running PyTorch workloads on AWS Trainium chips Microsoft targets the 'agentic insider' with new controls for rogue AI agents Nvidia, Dell back $200M round for AI model developer doubleAI LATEST STORIES Dream Labs raises $90M to help build more intelligent robots with simple video feeds EMERGING TECH - BY MIKE WHEATLEY . 8 HOURS AGO Trump administration orders ban on DeepSeek in federal agencies AI - BY JAMES FARRELL . 9 HOURS AGO AI work ass

匹配字段正文

t weeks after launching Amazon introduces toolkit for running PyTorch workloads on AWS Trainium chips Microsoft targets the 'agentic insider' with new controls for rogue AI agents Nvidia, Dell back $200M round for AI model developer doubleAI MiniMax launches M3.1 Flash model, says it cuts AI agent costs by 70% LATEST STORIES Dream Labs raises $90M to help build more intelligent robots with simple video feeds EMERGING TECH - BY MIKE WHEATLEY . 8 HOURS AGO AI work assistant startup Town secures un

匹配字段正文

arms can now be bought for the same price as a high-end laptop. “What they need now is intelligence,” she continued. “This team did some of the defining work in robot learning at Nvidia, and now anyone with an API key can build on it.” Photos: Dream Labs A message from John Furrier, co-founder of SiliconANGLE: Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create

匹配字段正文

rates. Those projects are all still believed to be on track. However, the future of the Loughton facility is now up in the air. Nscale originally announced plans to install 23,040 Nvidia graphics processing units at the site by the first quarter of 2027, which would have provided an initial 50 megawatts of AI capacity. It also had plans to expand that to 90 megawatts if needed. $MSFT Microsoft has canceled an AI data center deal with Nvidia-backed Nscale in the UK 🤖❌ The facility was expected to

匹配字段正文

n. Weiterlesen nach der Anzeige Die Financial Times wiederum zitiert in ihrem aktuellen Bericht als prominentes Beispiel für einen Neocloud-Anbieter das Unternehmen Nscale, an dem Nvidia maßgeblich beteiligt ist. Das Unternehmen soll bis August 2026 nur 25.000 von geplanten 500.000 KI-Chips online gehabt haben. Die Hoffnung ist, wie bei vielen KI-Anbietern, auch für Nscale ein Börsengang: Mit einer Bewertung von 35 Milliarden US-Dollar rechnet das Unternehmen. Mit OpenAI hat bereits das Schwerge

匹配字段正文

超过半数采用双方组合方案。以InfiniBand为例，Mellanox的InfiniBand高速互联技术占据市场超过70%的份额，谷歌、亚马逊、微软都是它的客户。 通过这笔收购，英伟达打通了NVLink（板内scale-up）+ InfiniBand/Spectrum（集群scale-out）全私有链路。黄仁勋当时表示：“通过融合Mellanox的技术，新的NVIDIA将拥有从AI计算到网络的端到端技术，以及从处理器到软件的全堆栈产品”。 这一阶段的本质是全封闭绑定体系：客户集群必须全套采购英伟达GPU+网络设备，性能极致但兼容性为零，客户无替代选项、议价权极低。 这套模式在大模型爆发早期红利巨大，但其瓶颈也逐步显现。陆续落地随着AI芯片国产化、AMD、英特尔持续突围、云厂商自研芯片爆发，叠加AI算力需求爆发带来的产能、交期紧张，单一供应商产能短缺、交期不稳，客户对于异构算力集群的需求快速提升，多芯片混插逐渐成为云厂商刚需，封闭生态的全栈绑定模式，越来越难以匹配行业客户诉求。 Token Fabric开启互联新范式 在此趋势下，英伟达的战略思维正在发生根本性升级。它不再试图阻止竞品入场

匹配字段正文

z umgesetzt ist dieser Konsens noch nicht. Auf der Linux-Kernel-Mailingliste ist ein entsprechender Patch indes schon eingereicht. Sasha Levin, Distinguished Software Engineer bei Nvidia und Co-Maintainer der Stable- und LTS-Kernel, beantragte dort am 9. Oktober, die Tag-Vorgabe aus der Dokumentation zu streichen. Der zuständige Maintainer Jonathan Corbet hat den Patch inzwischen in den Entwicklungszweig docs-next übernommen, in den Mainline-Kernel muss er es aber noch schaffen. Die Chancen dafü

匹配字段正文

more information and can publish without approval, while the top tier adds METR’s employee-level access. Thursday’s announcement makes good on a promise Anthropic CEO Dario Amodei made in a September essay , where he proposed embedding outside evaluators inside AI labs as a first step toward slowing the technology’s development. OpenAI quickly pledged the same , and both companies, along with Google, Meta, Nvidia, and xAI, have since signed a White House accord committing to independent review.

匹配字段正文

ucture needed to develop the technology. Stocks in some US technology companies dropped on Wednesday over concerns around investments in AI. Shares in chip makers - including AMD, Nvidia and Intel - fell sharply. In Asia, shares in Japanese tech investment firm Softbank were down by more than 9% on Thursday. South Korean semiconductor giants also dropped, with Samsung falling by 9.5% and SK Hynix 11% lower. Shares have been falling this week after remarks by Hock Tan, the chief executive of US t

匹配字段正文

stigators, but it had a massive impact on tech stocks. The Nasdaq index fell 1.25% and the shares of many individual firms, including Oracle Corp., CoreWeave Inc., Intel Corp. and Nvidia Corp., all fell more. Their fortunes are tightly linked to those of the AI giants, given that many have signed massive multiyear, multibillion-dollar contracts with them. Holger Mueller of Constellation Research said the number thrown by Anthropic at investors is all about making it look good ahead of its IPO, b

匹配字段正文

ctricity shortage. There are already signs investors are questioning how much money AI infrastructure companies are asking for. Firmus, an Australian data center company backed by Nvidia, withdrew its planned IPO on Oct. 9 after weak investor demand for its marketed share prices, according to Bloomberg. That doesn’t necessarily mean demand for AI computing is weakening. Intel CEO Lip-Bu Tan, who shared the panel with Tan at the conference, said hardware could remain in short supply for another t

匹配字段正文