OpenBMB@OpenBMB · 6月1日53

Just a quick reminder: Build Small Hackathon sign-up closes on June 3! 🏆Total cash prizes: ~$40K $10K @OpenBMB Special Awards for the best projects using #MiniCPM models. Don’t miss out! Register now：https://huggingface.co/build-small-hackathon We highly recommend using MiniCPM models — they’re small, powerful, and perfect for this hackathon. Let’s build something awesome together!🤗🚀

译快速提醒：Build Small黑客松报名将于6月3日截止！ 🏆总现金奖金：约4万美元 其中1万美元为@OpenBMB特别奖，授予使用#MiniCPM模型的最佳项目。 别错过！立即注册：https://huggingface.co/build-small-hackathon 我们强烈推荐使用MiniCPM模型——它们小巧、强大，非常适合本次黑客松。 让我们一起构建精彩项目！🤗🚀

Nathan Lambert@natolambert · 6月1日58

Nvidia is the tip of the spear for the growing US open model efforts. Is a long time coming, but obvious when a 550B model drops that more people will take notice. Many people won’t realize the valuable training data they release too.

译英伟达是美国日益壮大的开源模型努力的先锋。 虽然酝酿已久，但当一个550B模型发布时，更多人会注意到这一点，这是显而易见的。 许多人不会意识到他们发布的训练数据也很有价值。

Alibaba Cloud@alibaba_cloud · 6月1日58

Alibaba Cloud recently showcased our latest milestones at SAP Sapphire Madrid 2026, further deepening our global alliance. We are scaling 16TB RAM re8 ECS instances for SAP HANA and integrating Qwen LLM & Model Studio into SAP to power enterprise intelligence.

译阿里云近日在SAP Sapphire Madrid 2026上展示了最新里程碑，进一步深化我们的全球联盟。我们正在扩展用于SAP HANA的16TB RAM re8 ECS实例，并将通义千问大语言模型与Model Studio集成到SAP中，以赋能企业智能化。

Alibaba Cloud@alibaba_cloud · 6月1日37

Ready to race the 24-hour clock? 🎬 Join the 24H AI Film Hackathon at AI Film Fest Monaco 2026 (June 9-10), powered by Alibaba Cloud. Create your AI short film on-site to win credits (3000 USD) and main-stage screenings. Secure your Hackathon spot 👇https://int.alibabacloud.com/m/1000413821/

译准备好与24小时时钟赛跑了吗？🎬 加入2026摩纳哥AI电影节（6月9-10日）的24H AI Film Hackathon，由阿里云支持。现场创作你的AI短片，赢取积分（3000美元）和主舞台展映机会。 立即报名 👇https://int.alibabacloud.com/m/1000413821/

Alibaba Cloud@alibaba_cloud · 6月1日44

Creativity is entering its next wave — and we're fueling it. Meet Alibaba Cloud at @upscaleconf by @magnific in San Francisco, June 3–4, where artists, technologists, and storytellers from around the world come together to push what's possible with AI. Come create with us. ✨ 📍 The Midway, San Francisco 🗓 June 3–4, 2026

译创意正迎来下一波浪潮——我们正为其注入动力。 在旧金山由@magnific举办的@upscaleconf大会上与阿里云会面，6月3-4日，来自世界各地的艺术家、技术人员和故事创作者将齐聚一堂，共同探索AI的无限可能。 来与我们一起创造吧。✨ 📍 The Midway, 旧金山 🗓 2026年6月3-4日

X.PIN@thexpin · 6月1日52

BREAKING: Unitree just passed its IPO review on the SSE STAR Market! The Listing Review Committee confirmed today (June 1) that Unitree fully meets all issuance, listing, and disclosure requirements. Big milestone for the company!

译突发：宇树科技刚刚通过了上海证券交易所科创板的IPO审核！ 上市审核委员会今日（6月1日）确认，宇树科技完全符合所有发行、上市和信息披露要求。这是公司的一个重要里程碑！

MiniMax (official)@MiniMax_AI · 6月1日57

excited to team up with @BAI_AGI 🤝 M3 is bringing the brains, now agents get the money rails to operate on🔥

译MiniMax宣布与去中心化金融基础设施项目B.AI达成合作。合作旨在为AI智能体提供经济运作基础。B.AI正在构建金融基础设施，使智能体能够在去中心化经济中进行交易、协调和运作。随着MiniMax在推理、编码、多模态理解和智能体工作流方面推动前沿，此次合作将加速AI原生应用和自主智能体的未来发展。

SenseTime@SenseTime_AI · 6月1日41

We're excited to co-organise the 𝗔𝗴𝗲𝗻𝘁 𝗙𝗼𝗿𝗴𝗲 #𝗔𝗜 #𝗛𝗮𝗰𝗸𝗮𝘁𝗵𝗼𝗻: 𝗦𝘂𝗽𝗲𝗿 𝗔𝗜 𝗘𝗱𝗶𝘁𝗶𝗼𝗻 in Singapore!🛠️ Participants will: ⚡ Build a working AI app in ONE day 🧰 Use real tools: #SenseNova U1 (that's us! 👋), Bright Data, Daytona, Kimi AI, Nosana and Terminal 3 🏆 Demo live & compete for sponsor credit prizes 🤝 Connect with builders, engineers & founders Details: 📅 June 13, 2026 📍 SMU School of Economics, Singapore The event is fully booked — but you can still join the waitlist 👇 🔗 https://luma.com/agentforgesg Also join SenseTime's community for more updates 👇 👾Discord: https://discord.gg/BuTXPHmQub

译商汤科技宣布联合举办Agent Forge AI黑客松（Super AI Edition），活动将于2026年6月13日在新加坡SMU经济学院举行。参与者将使用SenseNova U1、Bright Data、Daytona、Kimi AI、Nosana和Terminal 3等工具，在一天内构建一个可运行的AI应用，并进行现场演示与竞赛，奖品为赞助商积分。该活动已满员，但可加入等待列表。

SenseTime@SenseTime_AI · 6月1日29

We're excited to co-organise the 𝗔𝗴𝗲𝗻𝘁 𝗙𝗼𝗿𝗴𝗲 #𝗔𝗜 #𝗛𝗮𝗰𝗸𝗮𝘁𝗵𝗼𝗻: 𝗦𝘂𝗽𝗲𝗿 𝗔𝗜 𝗘𝗱𝗶𝘁𝗶𝗼𝗻 in Singapore!🛠️ Participants will: ⚡ Build a working AI app in ONE day 🧰 Use real tools: Bright Data, Daytona, Kimi AI, Nosana, Terminal 3 — and #SenseNova U1 (that's us! 👋) 🏆 Demo live & compete for sponsor credit prizes 🤝 Connect with builders, engineers & founders Details: 📅 June 13, 2026 📍 SMU School of Economics, Singapore The event is fully booked — but you can still join the waitlist 👇 🔗 https://luma.com/agentforgesg Also join our community for more updates 👇 👾Discord: https://discord.gg/BuTXPHmQub

译活动名为“Agent Forge AI Hackathon: Super AI Edition”，将于2026年6月13日在新加坡SMU经济学院举办。参与者需在一天内构建可运行的AI应用，可使用的工具包括Bright Data、Daytona、Kimi AI、Nosana、Terminal 3以及主办方SenseNova的U1模型。活动现场将进行项目演示并角逐赞助商提供的奖金，同时提供与开发者、工程师及创始人交流的机会。该活动名额已满，但仍可加入候补名单。

SemiAnalysis@SemiAnalysis_ · 6月1日49

NVIDIA is proud to announce their partnership with Dwarkesh Patel.

译NVIDIA很荣幸地宣布与Dwarkesh Patel达成合作。

Runway@runwayml · 6月1日69

Introducing the Cosmos Coalition A new global initiative with NVIDIA and leading AI labs to build and open-source frontier world models for physical AI. Runway joins as a founding member, working alongside NVIDIA and a set of leading AI labs to build, share and accelerate world model research and development through a common open ecosystem.

译介绍Cosmos联盟 一项与NVIDIA及领先AI实验室合作的新全球倡议，旨在构建并开源用于物理AI的前沿世界模型。Runway作为创始成员加入，与NVIDIA及一系列领先AI实验室共同工作，通过一个共同的开放生态系统来构建、共享并加速世界模型的研究与开发。

Alibaba Cloud@alibaba_cloud · 6月1日44

Blooming Silent Dreams: Safeguarding Everyday Wonders with AI. Happy Children’s Day! 🎉 As the 1st AI-native project for children with disabilities, Qwen & Wan turn silent sketches from non-verbal kids into animations, ensuring voiceless dreams are heard. Watch👇

译绽放的无声梦想：用AI守护日常奇迹。 儿童节快乐！🎉 作为首个面向残疾儿童的AI原生项目，通义千问与万相将非语言儿童的无声涂鸦转化为动画，确保无声的梦想被听见。 观看👇

SemiAnalysis@SemiAnalysis_ · 6月1日68

BREAKING NEWS: JENSEN JUST ANNOUNCED MICROSOFT HAS FINISHED BRING UP ON THEIR FIRST RUBIN VR200 NVL72 RACK with their ODM partner, Foxconn.  Jensen also announced that wafer-level mass production has started. That being said, rack-level mass production has not started yet and remains in the engineering/quality sample stage.

译突发新闻：黄仁勋刚刚宣布，微软已与其ODM合作伙伴富士康完成了首台Rubin VR200 NVL72机架的验证工作。黄仁勋还宣布晶圆级量产已启动。但需注意，机架级量产尚未开始，目前仍处于工程/质量样品阶段。

MiniMax (official)@MiniMax_AI · 6月1日57

M3 live in Hermes Agent day zero 🔥 shows up in your model picker automatically shoutout @Teknium & the Nous team

译M3 在 Hermes Agent 首日上线 🔥 它会自动出现在你的模型选择器中 感谢 @Teknium 和 Nous 团队

小互@xiaohu · 6月1日53

Manus 引发的法规来了 出海可以，但是不能出卖... 投资者开展对外投资活动，不得出口、使用国家禁止出口的货物、技术、服务及相关数据，或者未经许可出口、使用国家限制出口的货物、技术、服务及相关数据； 不得以跨境派遣技术人员、组织人员赴其他国家（地区）工作、跨境提供技术指导、安排人员跨境培训等方式向其他国家（地区）转移国家禁止出口的货物、技术、服务及相关数据 或者未经许可向其他国家（地区）转移国家限制出口的货物、技术、服务及相关数据。

译中国发布新规，明确规范AI等技术的对外投资与跨境转移。规定要求，投资者在开展对外活动时，不得出口、使用或转移国家禁止出口的货物、技术、服务及相关数据；未经许可，也不得转移国家限制出口的相关内容。这特别涵盖了通过跨境派遣技术人员、组织境外工作、提供技术指导及安排跨境培训等方式进行的转移。

MiniMax (official)@MiniMax_AI · 6月1日38

We're going LIVE tomorrow with @togethercompute 🔥. @zpysky1125 is pulling back the curtain on M3: sparse attention, 1M context, all of it. You don't want to miss this.

译我们将于明天与 @togethercompute 一起直播🔥。 @zpysky1125 将揭秘 M3：稀疏注意力、1M 上下文，所有细节。 你不想错过这场直播。

Greg Brockman@gdb · 6月1日57

OpenAI Robotics is making rapid progress towards building AI that can help people in the physical world. Apply now to join the team:

译OpenAI宣布其世界模拟研究项目已演进为OpenAI Robotics团队，并在机器人与机器学习的协同设计基础上取得快速进展。该团队短期目标是开发能支持技术工人建设未来基础设施的机器人，长期愿景是为每个人配备个人机器人。团队正在招聘全栈硬件、运营、系统及机器学习工程师，旨在共同编程和制造对社会有用的机器人。

Emad@EMostaque · 6月1日78

Sora team became robotics team

译OpenAI的Sora世界模拟研究团队已转型为OpenAI Robotics机器人团队。团队由Aditya Ramesh领导，核心理念是AI应能帮助人类的物理世界。短期目标聚焦于开发支持技能工人、建设未来基础设施的机器人；长期愿景是为每个人打造个人机器人。该团队进展迅速，其基础是机器人硬件与ML研究的协同设计，目前正在招聘全栈硬件、系统及ML工程师。

Sam Altman@sama · 6月1日83

OpenAI Robotics is hiring, looking for exceptional full-stack hardware, ops, systems, and ML engineers to help us program and manufacture robots that are useful for society. AI should be able to help people in the physical world. In the short term, we are focused on robots to support skilled workers to build our future infrastructure; in the long term, we imagine everyone having a personal robot doing anything they need. Our world simulation research program, led by Aditya Ramesh (@model_mechanic), has evolved over the past year into OpenAI Robotics. Progress is rapid, and based on a foundation of co-design between robotics hardware and ML research. If you love working hands-on across the robotics stack and want to build the future, please consider joining us. Send an email with your background and evidence of exceptional accomplishment to: robotics-recruiting@openai.com

译OpenAI宣布成立OpenAI Robotics团队，并开始招聘全栈硬件、系统及ML工程师，以编程和制造能服务社会的机器人。该项目由Aditya Ramesh领导，其世界模拟研究计划已演变为机器人研究，强调硬件与ML研究的协同设计。短期目标是支持技术工人构建未来基础设施，长期愿景是为每个人提供个人机器人。

🚨 AI News | TestingCatalog@testingcatalog · 5月31日57

Anthropic is planning to further expand into the consumer and bioscience sectors. The biggest things to watch for 👀 - Conway agent - Orbit assistant - Knowledge-based memory - Multilingual Voice Mode - Operon for bioscience researchers and more! Which one do you think will drop next?

译Anthropic计划进一步扩展至消费与生物科学领域，并预告了多款即将推出的产品，包括Conway agent、Orbit assistant、知识记忆、多语言语音模式以及面向生物科学研究的Operon。引用观点指出，Anthropic选择先聚焦编程，但随着Claude的智能提升，其应用将扩展到人类智能能发挥作用的各个领域。

Chubby♨️@kimmonismus · 5月31日75

Interesting updates on Apple AI: As Apple's WWDC lands next month, and the long-delayed Siri and on-device AI upgrades are expected to be the centerpiece: a smaller, distilled version of Google's Gemini running locally on iPhone silicon, pitched on privacy and lower token costs. Most of that stack is sourced from elsewhere. The local model is distilled from Gemini. Queries too heavy for the device route to Google Cloud (!), where Apple has now signed off on Nvidia's confidential-compute tech to process them. Apple is also reportedly hunting for small on-device-AI startups to speed up the model-shrinking work, having looked at Liquid AI among others. One quiet shift from the 2024 rollout: Apple promised then that anything leaving your iPhone would run on Apple silicon inside Private Cloud Compute. It couldn't get the full Gemini running there, so those queries now sit in Google Cloud. The Private Cloud Compute name is staying anyway. Via The Information

译苹果下月WWDC将重点展示延迟已久的Siri及设备端AI升级，核心是在iPhone芯片本地运行从Google Gemini蒸馏而来的更小模型，以强调隐私与降低token成本。但该技术栈大部分源自外部：本地模型由Gemini蒸馏，设备无法处理的复杂查询将路由至Google Cloud处理，并采用了Nvidia的机密计算技术。苹果据称正在寻觅小型设备端AI初创公司以加速模型缩减工作。此外，苹果2024年承诺的Private Cloud Compute方案原计划使用苹果芯片，因无法承载完整Gemini模型，现已部分转向谷歌云，但该名称仍将保留。

AYi@AYi_AInotes · 5月31日69

Damn，Anthropic破万亿了？！ 所有喊AI泡沫的人，今天集体沉默了：

译天啊，Anthropic破万亿了？！ 所有喊AI泡沫的人，今天集体沉默了：

宝玉@dotey · 5月31日27

昨天不重置明天要重置🙄

SemiAnalysis@SemiAnalysis_ · 5月31日61

BREAKING NEWS: COREWEAVE & DELL IS THE FIRST CLOUD TO ANNOUNCE THAT THEY HAVE RUBIN VR200 NVL72 WITH FULLY PASSING L11 DIAGS. Next Step is to get multiple racks burnin in a couple & do software level bringup like sglang, vllm, dynamo, etc.

译突发新闻：CoreWeave与Dell是首个宣布其Rubin VR200 NVL72已完全通过L11诊断的云服务商。下一步是获取多个机架进行数周的烧机测试，并完成软件层面的启动工作，如SGLang、vLLM、Dynamo等。

Replit ⠕@Replit · 5月31日27

Reimagining technology as a catalyst for human connection, @tigrisli designs installations, electronic hardware, and experiences that evoke curiosity and play. See her perfume installation, transforming memories into bespoke fragrance with AI in real time, on day two of Vibecon. NYC, June 17–18. Get your tickets at http://vibecon.ai

译@tigrisli 将技术重新构想为促进人际连接的催化剂，她设计装置、电子硬件和体验，以唤起好奇心与趣味性。 在 Vibecon 第二天，观看她的香水装置，该装置利用 AI 实时将记忆转化为定制香气。 纽约，6月17–18日。 购票请访问 http://vibecon.ai

Rohan Paul@rohanpaul_ai · 5月31日76

FT: SoftBank just pledged €75B to build Europe’s largest AI computing facility in France, turning cheap, stable nuclear-heavy power into the raw fuel for training and running bigger AI systems. The plan starts with €45B for 3.1GW of capacity in Hauts-de-France by 2031, then adds another 2GW, creating a 5GW AI complex roughly comparable to the power draw of a major city. Dunkirk is the anchor site, with Schneider Electric expected to help build an AI infrastructure and robotics manufacturing hub around the project. The weak point is financing, because industry estimates put 1GW of AI infrastructure near $50B, meaning SoftBank will probably need heavy project debt and outside partners. --- ft .com/content/1022f9bd-5b6d-44a5-9303-c8b05b8c6463?syn-25a6b1a6=1

译软银宣布在法国投资750亿欧元，旨在打造欧洲规模最大的AI算力设施。该计划旨在利用当地稳定廉价的核电。项目分阶段推进，第一阶段投资450亿欧元，目标是到2031年在法国北部-加来海峡大区建成3.1GW容量，随后再增加2GW，最终形成5GW的AI综合体。敦刻尔克是核心选址，施耐德电气预计参与建设AI基础设施及机器人制造中心。融资是潜在挑战，行业估算显示1GW的AI基础设施成本约为500亿美元，因此软银可能需要依赖大量项目债务和外部合作伙伴。

Chubby♨️@kimmonismus · 5月31日71

Fortune reports Microsoft is building a "super app" to unify its scattered Copilots. Under 4.5% of 450 million Microsoft 365 seats pay for Copilot. Around 20 million, out of nearly half a billion. The app is pitched as a fix for fragmentation. The open question is whether fragmentation is really the problem, or whether no single interface moves a number that low. GitHub Copilot is the exception at 4.7M paid, and it's the one now facing real pressure from Cursor and Claude Code. Source: Fortune

译据《财富》报道，微软正构建一个“超级应用”，以统一其分散的Copilot产品。此举的背景是付费率低迷：在Microsoft 365近5亿席位中，仅约2000万（不到4.5%）付费使用Copilot。GitHub Copilot是例外，拥有470万付费用户，但正面临来自Cursor和Claude Code的竞争压力。核心疑问在于，产品分散是否是付费率低的主因，抑或单一界面本身难以驱动用户大规模付费。

OpenAI Developers@OpenAIDevs · 5月31日59

🧵 Our Voice Hack Night finalists are here. 4 projects. 6 hours. Realtime voice agents in real-world builds. Now it’s your turn to vote for your favorite. We’ll announce the winner on Monday. https://cerebralvalley.ai/e/openai-voice-hack-night/hackathon/gallery

译🧵 我们的语音黑客之夜决赛作品已揭晓。 4个项目。6小时。真实场景中的实时语音智能体现已构建完成。 现在轮到你投票选出你最喜欢的作品了。我们将于周一公布获胜者。 https://cerebralvalley.ai/e/openai-voice-hack-night/hackathon/gallery

Eric@ericmitchellai · 5月30日37

OpenAI is famous for leaks but man to just go out there and tell people we will train more models Huge alpha in following this guy

译OpenAI 以泄露闻名，但直接站出来告诉大家我们会训练更多模型 关注这个人能获得巨大优势 [引用 @thsottiaux]：当我们从 GPT-5.0 -> GPT-5.1 -> ... -> GPT-5.5 时，版本号的递增伴随着能力提升和 token 效率改进（这转化为速度提升）。GPT-5.5 是我们迄今最好的模型。 这是我们希望继续的简单策略。

StepFun@StepFun_ai · 5月30日30

Glad it was smooth. Now comes the interesting part!

译很高兴一切顺利。现在有趣的部分来了！

Rohan Paul@rohanpaul_ai · 5月30日74

The information: Meta is preparing its biggest AI wearable push yet, with a AI pendant, more AI glasses, and a business service called Wearables for Work. Meta’s bet is that the next AI interface is not a chat box, but a sensor-rich device tied to an assistant that can remember meetings, summarize conversations, answer visual questions, and trigger actions. The reported target is huge: 10M wearable sales in the second half of 2026 and 6.8M monthly active wearable users by year-end. The software layer may matter more than the hardware, because Meta AI Assistant, Hatch, subscriptions, and wearable apps turn a device sale into recurring AI revenue. The pressure is obvious: Reality Labs posted a $4.03B operating loss on only $402M revenue in Q1-26, so Meta needs wearables to become a platform, not another expensive gadget line. --- theinformation .com/briefings/meta-plans-ai-pendant-part-ambitious-wearables-expansion

译Meta正准备迄今规模最大的AI可穿戴设备推进，包括AI项链、更多AI眼镜以及企业服务“Wearables for Work”。其押注下一代AI交互界面不是聊天框，而是具备丰富传感器、能记住会议、总结对话、回答视觉问题并触发操作的AI助手设备。报道的销售目标宏大：2026年下半年销量目标1000万台，年底月活用户目标680万。软件层被视作关键，可将设备销售转化为持续性AI收入。此举背后的财务压力明显：Reality Labs在2026年第一季度录得40.3亿美元运营亏损，营收仅为4.02亿美元，因此Meta亟需将可穿戴设备发展成一个平台，而非又一条昂贵的硬件产品线。

Chubby♨️@kimmonismus · 5月30日63

A team of former DeepMind researchers just raised $50M to build an AI lab built around recursive self-improvement at the level of the whole research organization, not only a single model. Index and Radical co-led, NVIDIA's venture arm is in, and angels like Dwarkesh Patel, Thomas Wolf and Max Jaderberg are on the cap table. The founders have the track record to back it up. Louis Kirsch comes out of the Schmidhuber lineage on self-improving systems. Edward Hughes has argued that open-endedness is essential for artificial superhuman intelligence. Tantum Collins worked on AI policy in the Biden White House. Their idea is simple and big at the same time. Today's models are great at answering questions, but real discovery also depends on knowing which questions are worth asking. Inherent wants AI that works right next to humans inside that loop, as a collaborator and not only a tool. They call it living within the experiment. They also set it up as a Public Benefit Corporation, so the mission is written into the company from day one. This is the direction a lot of us have been hoping for, and one of the more credible attempts at recursive self-improvement I've seen so far. Really excited for it.

译由前 DeepMind 研究员创立的 AI 实验室 Inherent 完成了 5000 万美元的种子轮融资，由 Index Ventures 和 Radical 共同领投，NVIDIA 旗下风投部门 NVentures 参投。创始团队包括 Louis Kirsch、Edward Hughes 和 Tantum Collins。该公司旨在构建能够主动发现新知识的 AI 智能体，其核心理念是实现整个研究组织的“递归自我改进”，使 AI 成为人类研究中的协作伙伴。Inherent 被定位为一家公共利益公司，总部位于伦敦。

凡人小北@frxiaobei · 5月30日26

罢了，错信了 Claude，新账号且用且珍惜。

Rohan Paul@rohanpaul_ai · 5月30日63

Reuter: ByteDance is building its own AI data-center CPUs because running agents at TikTok scale now depends on scarce server processors, not only Nvidia GPUs. inspired by Groq's "language processing units," they are testing both Arm and RISC-V, which lets it compare a mature commercial design against a more controllable open instruction set before mass production. The market is seeing a 10%-35% quarterly CPU price increases and long supply delays, hence making an in-house silicon is now cost and supply-chain move, not just a prestige project. So ByteDance wants to both reduce dependence on restricted foreign AI hardware and make inference cheaper per query. The deeper shift is that AI agents is now turning CPUs into strategic chips. A gentic inference stresses CPUs much more because one user request can trigger many smaller steps: retrieve files, call a tool, query a database, run a model, check the answer, call another model, send data across servers, and manage memory. However, ByteDance does not seem to have in-house chip design teams and is reportedly relying on several external partners, who are also expected to handle the actual silicon manufacturing. --- reuters .com/world/china/bytedance-developing-custom-cpu-chips-support-ai-rollout-sources-say-2026-05-28/

译路透社报道称，字节跳动正开发自研数据中心CPU芯片，以支持TikTok规模的AI智能体运行。此举受Groq的“语言处理单元”启发，旨在应对当前服务器处理器短缺问题。公司正在测试Arm和RISC-V两种架构，以比较成熟商业设计与可控开放指令集。由于CPU价格季度性上涨10%-35%且供应链延迟，开发自研芯片已成为一项成本与供应链策略，旨在减少对受限外国AI硬件的依赖并降低单次查询推理成本。AI智能体的推理对CPU依赖远大于传统模型，因单个用户请求可能触发多个步骤。据报道，字节跳动可能依赖外部合作伙伴进行芯片设计与制造。

Rohan Paul@rohanpaul_ai · 5月30日63

Reuter: Japanese banks are getting early access to OpenAI’s newest model for security testing, which is believed to be on a par with that of rival Anthropic's Claude Mythos. There were also reports that Japan's three largest banks are expected ​to gain access to Anthropic Mythos. Adoption of the new models is expected to help stop new types of cyberattacks. These 2 “frontier AIs” are said to detect vulnerabilities far better than earlier models. --- reuters .com/world/asia-pacific/openai-gives-japan-banks-access-latest-model-japans-finance-minister-says-2026-05-29/

译日本银行已获得OpenAI最新模型的早期访问权限，用于网络安全测试。据报道，该模型性能据称与竞争对手Anthropic的Claude Mythos相当。同时，日本三大银行预计将获得Anthropic Mythos的访问权限。这两款前沿AI模型被期望能帮助阻止新型网络攻击，因为它们在检测漏洞方面远优于早期模型。

Chubby♨️@kimmonismus · 5月30日56

A 95-minute AI-generated action movie just screened at Cannes’ Marché du Film. It was made in just two weeks for around $500,000, with most of the budget going into compute, proving that AI filmmaking is moving from demo clips to full-length productions. And that's special. They managed to produce the film's length (the clips created are usually very short) while staying within a very low budget.

译一部95分钟的AI生成动作片刚刚在戛纳电影市场放映。 它仅用两周时间、约50万美元预算制作完成，大部分预算用于算力，证明AI电影制作正从演示片段转向完整长片。 这很特别。他们设法制作了电影长度的作品（通常生成的片段非常短），同时保持了极低的预算。

SemiAnalysis@SemiAnalysis_ · 5月30日76

BREAKING NEWS: JAX NVIDIA GPU & XLA: GPU's biggest customer just announced that they have dropped JAX GPUs and would rather vibe code an C training framework with Grok Build. Reportedly xAI's JAX stack had MFU lower than 10%. NVIDIA JAX team entire main 996 focus for the past 2 years was to support xAI yet has failed, I guess they can rest & vest now.

译突发新闻：JAX NVIDIA GPU & XLA：GPU最大客户刚刚宣布已放弃JAX GPU，宁愿用Grok Build“氛围编程”一个C训练框架。据报道xAI的JAX堆栈MFU低于10%。NVIDIA JAX团队过去两年全部主力996专注于支持xAI却失败了，我想他们现在可以休息并兑现期权了。

Tibo@thsottiaux · 5月30日38

I looked at a number today on a codex dashboard and it made me happy. More news about the number soon. 👀 Thanks to everyone who keeps adopting codex. We are still early. So early.

译我今天在Codex仪表盘上看到了一个数字，它让我很开心。关于这个数字的更多消息即将公布。👀 感谢所有持续采用Codex的用户。我们仍处于早期阶段。非常早期。

Alibaba Cloud@alibaba_cloud · 5月30日64

UEFA and UC3 have announced Alibaba Group as global AI, Cloud Computing and E-Commerce Partner for UEFA EURO 2028, as well as UEFA men’s club competitions from 2027 to 2033. Alibaba Cloud and Qwen will leverage its leading artificial intelligence (AI) technologies, cloud computing capabilities and global e-commerce platform to deliver intelligent innovations across UEFA’s flagship competitions and create a more immersive fan experience worldwide.

译阿里云和Qwen成为UEFA官方独家AI、云计算与电子商务合作伙伴，合作期覆盖2027/2028赛季至2032/2033赛季的UEFA男子俱乐部赛事，以及UEFA EURO 2028。阿里巴巴集团主席蔡崇信表示，将投入云计算、全栈AI及全球电商平台能力，支持赛事运营。合作将利用Qwen大语言模型部署先进AI技术，增强球迷互动与媒体内容体验，并依托阿里云基础设施打造全球沉浸式观赛体验。

SemiAnalysis@SemiAnalysis_ · 5月30日67

TRUTH SOCIAL: NVLink multicast is not supported on Blackwell "Confidential Computing" leading to 61% performance regression on SGLang Qwen3.5 397B according to @verdacloud 's recent github ticket. NVIDIA's  "Confidential Computing" is complete slop as in addition Hopper's  confidential computing had fully unencrypted NVLink according to NVIDIA's own "NVIDIA Secure AI with Blackwell and Hopper GPUs" Whitepaper.

译TRUTH SOCIAL：根据@verdacloud最近的GitHub工单，NVLink多播在Blackwell“机密计算”上不被支持，导致SGLang Qwen3.5 397B性能下降61%。NVIDIA的“机密计算”完全是垃圾，此外根据NVIDIA自己的《NVIDIA Secure AI with Blackwell and Hopper GPUs》白皮书，Hopper的机密计算也存在完全未加密的NVLink。