Berlin, you look good in CUDA. 💚
Berlin, you look good in CUDA. 💚 #NVIDIAGTC is coming to town. We thought we’d say hello from the TV Tower. Spot it? Tag @NVIDIA and show us your view. #FestivalOfLights https://t.co/Wy0oiBisAb
#英伟达
X（推特）上的财经、科技与宏观账号。按时间排，不论是否入选精选。
X·科技公司AI 评分1313 X·财经博主AI 评分6060 Hut 8与英伟达、Lambda的数据中心租赁与GPU部署安排
Hut 8建设数据中心并与英伟达签订15年租约，英伟达作为租户以其信用支持建设融资，随后将空间交由Lambda安装GPU并出租。引用的SemiAnalysis称，该安排涉及约115K台Rubin GPU销售、Lambda收入约10%的分成，以及一个704 MW供电园区第7至第15年的空间；互联排队时间为4至7年。
X·科技公司AI 评分1212 They show up for us. Who shows up for them?
They show up for us. Who shows up for them? Meet Wenyi, Founder of Kuddo Health. Her experience with therapy sparked a mission to support clinicians, with help from NVIDIA GPU–accelerated AI. Watch her story. #WorldMentalHealthDay https://t.co/Kee5GEGHej
X·宏观AI 评分4141 美国贸易逆差中的“AI”部分增长，主要体现为自墨西哥进口增加
美国贸易逆差中的“AI”部分正在增长，主要体现为自墨西哥进口增加；台湾及其他代工厂也已大规模转向墨西哥进行最终组装。Foxconn位于墨西哥哈利斯科州埃尔萨尔托、组装英伟达 NVL72 服务器机架的工厂，目前供应的商品接近美国商品进口总额的 1%。
X·财经博主AI 评分5050 英伟达 Vera Rubin 服务器启动生产，Blackwell GB300 和 B300 服务器量产推动 Wistron 营收创新高
英伟达 Vera Rubin 服务器启动生产，Blackwell GB300 和 B300 服务器量产，推动 AI 服务器厂商 Wistron 营收创纪录新高。AMD CEO Lisa Su 近期访问 Wistron 位于 Zhubei 的新厂 A1，显示 Helios 服务器业务延续强势。
X·宏观AI 评分88 Tensor10 (see what I did there having added AMD) chart/table
Tensor10 (see what I did there having added AMD) chart/table https://t.co/JidNwVlRhK
X·科技公司AI 评分1313 How H-Company Optimizes VLM Serving for Computer Use Agents With NVIDIA Dynamo
How H-Company Optimizes VLM Serving for Computer Use Agents With NVIDIA Dynamo https://t.co/gixR8bJLTj
X·宏观AI 评分1919 昨天视频下面留言：“SP500的EPS最大的贡献者就是NVDA和MU，占了1/3，所以SP500的估值并不能简单说是便宜”
昨天视频下面留言：“SP500的EPS最大的贡献者就是NVDA和MU，占了1/3，所以SP500的估值并不能简单说是便宜” 下面是标普等权，15.9倍，也低于今年3月份，比上年的3月，15倍，要高。高了大概6%。 估值在低位的结论不变，我就是一个在数据上非常执着的杠精 https://t.co/jIByYRLd9o
X·宏观AI 评分2121 Many nasty replies to my piece yesterday on flatlining model revenue b/c they were highly confident…
Many nasty replies to my piece yesterday on flatlining model revenue b/c they were highly confident OpenAI ARR surged to 70bln based based on an unsourced leak. Lots mocking the use of @tickerplus as an estimate. Turns it was... 50bln. smh https://t.co/MQWQAAMi79 https://t.co/X47HrfOpom 引用 @BobEU
X·宏观AI 评分3333 FT: OpenAI ARR比先前市场预估的少了200亿，市场下跌
FT: OpenAI ARR比先前市场预估的少了200亿，市场下跌 这种新闻怎么理解？ 1. 首先ARR不是营收，你可以拿1个月去年化，1周去年化，或者1天去年化，除非内部改标准（比如上年用1个月年化，今年用1天年化），那就是很严重的欺诈，不然只要标准一样没问题，看的是同比增速。当然上市后有监管的衡量方法 2.
X·科技公司AI 评分4141 Anthropic承诺向Genesis Mission投入1.5亿美元，并向15个以上联邦机构提供Claude和技术支持
Anthropic承诺向Genesis Mission投入1.5亿美元，以加大对科学研究和技术发现的投入。公司还将向15个以上联邦机构提供Claude和技术支持。
X·科技公司AI 评分2424 Congratulations to our CEO on receiving the National Medal of Science for advancing GPU computing…
Congratulations to our CEO @JensenHuang on receiving the National Medal of Science for advancing GPU computing to power scientific breakthroughs. And to fellow honorees Elon Musk, Lisa Su, Michael Dell, Satya Nadella, and Sergey Brin. 引用 @RapidResponse47：.@POTUS presents the National Medal of Scie
X·科技公司AI 评分2121 And Grok only gets better from here 🚀 🚀
And Grok @Bot only gets better from here 🚀 🚀 You could build an entire company made of Grok Bots! 引用 @MiaAI_lab：Grok @bot might be the most important AI release of 2026. I'm just blown away by what it can do now. A few days ago it felt like a niche toy and I didn't use it much... but now?! It's
X·科技公司AI 评分1313 Congratulations to Sergey, Jensen, Lisa, Michael and Satya!
Congratulations to Sergey, Jensen, Lisa, Michael and Satya! 引用 @RapidResponse47：.@POTUS presents the National Medal of Science to Sergey Brin, @JensenHuang, @elonmusk, and @LisaSu, and the National Medal of Technology and Innovation to @MichaelDell and @satyanadella. https://t.co/8IGZV6s9Of
X·财经博主AI 评分3838 Gigabrains help me out:
Gigabrains help me out: GS puts required annual hyperscale AI revenue at $526bn, based on average annual capex across 2026 and 2027 and a 20% hyperscaler ROIC. I assume OpenAI and Anthropic have similar ARR, GW capacity and inference share, with $20bn of revenue per blended GW
X·科技公司AI 评分3333 SynthID Detector is now available to everyone. 🌐
SynthID Detector is now available to everyone. 🌐 Check whether online content was generated using @GoogleAI, or with tools from our industry partners – including @OpenAI, @NVIDIA, Kakao and coming soon, @Apple. Try it out → https://t.co/cPW2aNnqr6 https://t.co/TGYA0uZ0dx
X·科技公司AI 评分3030 Grok will use whatever achieves the best outcome for users.
Grok @Bot will use whatever achieves the best outcome for users. Simple questions will route to small, fast models. Questions with complex answers will route to large models. 引用 @Scobleizer：Grok goes open. Holy shit.
X·科技公司AI 评分4747 SynthID Detector 现已向所有人开放
SynthID Detector 现已向所有人开放，可用于检查在线内容是否由 Google AI 或行业合作伙伴的工具生成。合作伙伴包括 OpenAI、NVIDIA 和 Kakao，Apple 即将加入；NVIDIA 表示，任何人都可以上传由 NVIDIA Cosmos 生成的内容，检查其是否来自 NVIDIA 模型。
X·科技公司AI 评分3434 The new Type 01 is here, powered by NVIDIA.
The new @Jaguar Type 01 is here, powered by NVIDIA. Under its stunning design is NVIDIA Hyperion, the computer and sensor platform that takes in everything the car sees and senses, then helps it make smart decisions in real time. It’s paired with NVIDIA Halos, a safety system https://t.co/3Ra8YQih
X·科技公司AI 评分1616 However, most requests are pretty simple and will be handled by a lightning-fast version of Grok…
However, most @Bot requests are pretty simple and will be handled by a lightning-fast version of Grok 4.8 when that comes out. Operating principle is to give Grok Bot users the best possible combination of speed & intelligence. 引用 @minchoi：This is wild... Grok Bot will now pick Claude Opus 5.5,
X·科技公司AI 评分1818 An AI agent makes a mistake early in a task, then keeps going in the wrong direction.
An AI agent makes a mistake early in a task, then keeps going in the wrong direction. Our researchers built PivotOPD to teach agents how to avoid those mistakes and recover when they happen. During training, a teacher model shows the agent a better action and how to get back on https://t.co/7hMLMdE
X·财经博主AI 评分4545 Grok Bot 改用灵活模型部署，SpaceX 将按任务选择后端模型
Grok Bot 将采用灵活的模型部署方式。SpaceX 表示，今后会针对不同任务选用最合适的后端模型，包括 Claude Opus 5.5、MidJourney、Suno 等领先 API，以提供更好的结果。
X·财经博主AI 评分3636 Guess we’ll see how open GPU financing markets remain with how this deal does
Guess we’ll see how open GPU financing markets remain with how this deal does PIMCO apparently interested 引用 @business：The AI arms race is running on debt. SpaceX is looking to borrow $40 billion for Nvidia chips, while Ray Dalio warns the boom is a “classic bubble” near its breaking point: Evenin
X·财经博主AI 评分5353 英伟达支持的 Project Southgate 计划到 2028 年在澳大利亚部署最高 1.6GW AI 算力
英伟达支持的 Project Southgate 去年 10 月公布计划，拟到 2028 年在澳大利亚部署最高 1.6GW AI 算力；一年后，已部署规模约为 42MW，CDC 开发合作关系结束。原文称，该项目规划利用可再生能源提供主权 AI 算力，并由 CDC 提供数据。
X·宏观AI 评分5858 作者称黄仁勋认为MRVL未来达到万亿美元市值的概率较高
作者称，黄仁勋认为MRVL未来达到万亿美元市值的概率较高。引用内容称，MRVL宣布与谷歌于7月29日达成芯片产品开发协议，涵盖AI、存储和网络芯片领域，协议潜在总值为1200亿美元；MRVL还向谷歌发行最多可购买5897万股股票的认股权证，行权价格为每股206.58美元。协议签署后的第一年内，每季度将有136.0867万份认股权证股份归属。
X·宏观AI 评分1515 $GOOG 收购 $CEG，不是在卖飞就是在卖飞的路上，等着 $VST 弥补我的损失！
$GOOG 收购 $CEG，不是在卖飞就是在卖飞的路上，等着 $VST 弥补我的损失！ 能源 = 电 = 数据中心 = TOKEN = 智力 = 营收
X·科技公司AI 评分1616 Some telecom network problems don’t have a fix in the runbook.
Some telecom network problems don’t have a fix in the runbook. AI agents can investigate, compare fixes and test them in simulation. NVIDIA Nemotron helps agents reason through problems, while NVIDIA OpenShell controls what they can access and do. See how: https://t.co/19eWA7PQZ2
X·科技公司AI 评分3333 In our 2026 telecom AI survey:
In our 2026 telecom AI survey: • 89% of respondents said open source models and software are important to their AI strategy. • 66% said their company is actively using AI, up from 49% last year. Read the report: https://t.co/qxRZ6dxG1O https://t.co/ak66W8E4Xs
X·科技公司AI 评分2828 Telcos need AI agents that understand their networks.
Telcos need AI agents that understand their networks. We provide the full stack to turn open models into specialized agents while keeping operators in control. @ATT, @SoftBank and @Indosat are adopting open models for use cases like autonomous network operations and adapting https://t.co/1fxnyZaVA
X·科技公司AI 评分1212 Grok can be your personal chief financial officer!
Grok @Bot can be your personal chief financial officer! 引用 @Teslaconomics：I literally hooked up ALL my finance accounts to Grok Bot. I really mean all of them: 1/ Chase 2/ Charles Schwab 3/ E*TRADE 4/ Fidelity 5/ Robinhood 6/ Wealthfront 7/ Apple Card 8/ 𝕏 Money FYI, this is a total of 30 acco
X·科技公司AI 评分1010 Evaluating Open Models with Artificial Analysis | Nemotron Labs
Evaluating Open Models with Artificial Analysis | Nemotron Labs https://t.co/PnydgvUsKM
X·财经博主同新闻AI 评分7676 Google与Constellation公布两项核电协议细节，合计签约3.6 GW
Google与Constellation公布两项协议细节，合计签约3.6 GW。第一项为890 MW新增核电容量，签订20年购电协议，并涉及伊利诺伊州、宾夕法尼亚州和新泽西州六处站点的11台机组增容，首项增容预计于2028年完成；第二项涉及2,700 MW，原文未提供协议细节。
同一新闻，精选展示《谷歌与Constellation Energy签署20年电力购买协议，拟为PJM新增890兆瓦核电装机》
X·财经博主AI 评分1616 Don’t these charts suggest that the monthly subs are still just heavily subsidized by the companies…
Don’t these charts suggest that the monthly subs are still just heavily subsidized by the companies compared to enterprise APIs and that’s unsustainable 引用 @kimmonismus：Holy sh*t, Anthropic’s subscriptions offer so much more value, even compared to the new and efficient GPT-6.1 Sol. It’s not even c
X·科技公司AI 评分1313 The model behind this result is now on 🤗
The model behind this result is now on @huggingface 🤗 Nemotron-Labs-3-Competitive-Coding is a competitive-programming specialist model based on Nemotron-3-Ultra. https://t.co/xEegK23pQ0
X·宏观AI 评分1818 EPS压制PE的美股，献给还在刻舟求剑，疑惑利率上涨，英伟达没像22年那种下跌却破新高的各位。
EPS压制PE的美股，献给还在刻舟求剑，疑惑利率上涨，英伟达没像22年那种下跌却破新高的各位。
X·财经博主AI 评分4444 Nvidia和博通给出未来12个月AI收入展望，合计约8000亿美元
Nvidia和博通给出的未来12个月AI收入展望合计约8000亿美元。MS认为，当前电力限制不会危及这一收入，但可能使业绩上行更难；两家公司合计占XPUs市场约90%。
X·财经博主AI 评分5858 Shanu Mathew以ERCOT约90GW峰值负荷对照UBS数据中心机架容量预测
Shanu Mathew称，UBS预计全球年度机架容量部署将从2026E的26.5 GW升至2030E的104.5 GW，其中50.5 GW将用于Nvidia。作者以ERCOT当前约90GW的峰值负荷作对照，称部署全部容量相当于满足ERCOT目前的峰值负荷；原文列出$NVDA、$GOOGL、$AMD、$MSFT、$META、$ORCL和$AMZN。
X·财经博主AI 评分2929 Per reports, both Microsoft and Meta with significant cuts. Given Anthropic’s revenue concentration…
Per reports, both Microsoft and Meta with significant cuts. Given Anthropic’s revenue concentration and contribution from top 2 customers (of which many people assume Meta was), this could be a meaningful drag on the trajectory of growth. Token cost optimization era has began in 引用 @amir：Microsoft
X·财经博主AI 评分2929 OAI and Ant are also at ~30-40% inference with the rest allocated for training
OAI and Ant are also at ~30-40% inference with the rest allocated for training It’s one of the most important assumptions in all of compute/tokenomics because it dictates how much compute is monetized and the monetization rate As these guys all scale from 1-2 gigs to 5 GW and 引用 @jukan05：Appare
X·财经博主AI 评分2121 I get the call out on aggressive valuations for the labs (which should be litigated in public and…
I get the call out on aggressive valuations for the labs (which should be litigated in public and we’ll see how the IPOs do) but comparing multiples or against historical revenue without accounting for growth (amongst the fastest we’ve ever seen in capitalism) and the rapid 引用 @craigweiss：if meta w