Thariq@trq212 · 37分钟前48

I've heard a lot of questions about Fable's availability on subscription plans. While it will come off subscriptions after July 7th, we aim to restore Fable as a standard part of our subscriptions as soon as capacity allows, as we mentioned in our original blog post.

译我听到很多关于Fable在订阅计划中可用性的问题。 虽然它将在7月7日后从订阅中移除，但正如我们在最初博客文章中提到的，我们旨在尽快在容量允许的情况下将Fable恢复为我们订阅的标准部分。

SemiAnalysis@SemiAnalysis_ · 58分钟前27

Meta Compute: Everyone Wants To Be A Cloud Zuck Takes Plan B? SpaceX 2.0, Bedrock 2.0, MSL Isn't Giving Up, Scaling RecSys by 10x... ClusterMAX ranking coming soon? https://newsletter.semianalysis.com/p/meta-compute-everyone-wants-to-be

译Meta计算：人人都想成为云 扎克伯格选择B计划？ SpaceX 2.0、Bedrock 2.0、 MSL并未放弃，将推荐系统扩展10倍……ClusterMAX排名即将推出？ https://newsletter.semianalysis.com/p/meta-compute-everyone-wants-to-be

Replit ⠕@Replit · 1小时前29

The installations, the workshops, the talks, the people. A few of the moments that made vibecon what it was. Thank you for building it with us. ⚡

译装置、工作坊、演讲、人群。这些瞬间塑造了vibecon。 感谢与我们一同打造这场盛会。⚡

Rohan Paul@rohanpaul_ai · 2小时前48

Palantir says some US government customers are moving sensitive AI work to Nvidia Nemotron open models. Nvidia says Nemotron can run in air-gapped environments, away from unsecured external networks. That setup matters for defense work, where data leaks can damage missions immediately. Karp framed Nemotron as equal or better for classified battlefield-style tasks, according to the report. The deeper change is that model quality is no longer the whole purchase decision. Agencies now weigh speed, price, sovereignty, audit trails, and operational control together. American open models give Washington a third path between closed labs and foreign stacks. --- breakingthenews. net/Article/Palantir-CEO:-Some-US-gov't-clients-using-open-source-AI/66625598

译Palantir CEO透露，部分美国政府客户正在将敏感AI工作迁移至Nvidia的Nemotron开源模型。Nemotron可在气隙环境运行，隔离于不安全的外部网络，适合对数据泄露敏感的国防任务。Karp称Nemotron在机密战场级任务上表现相当或更优。更深层变化在于，模型质量不再是唯一购买决策因素，机构现在综合权衡速度、价格、主权、审计追踪和运营控制。美国开源模型为华盛顿提供了介于闭源实验室和海外技术栈之间的第三条路径。

jason@jxnlco · 2小时前16

Come say hi!

译今天在 @aiDotEngineer 大会的 OpenAI 展台，下午 2 点将举办 @theo 的 AMA（有问必答）活动。现场还有抓娃娃机、奶茶和著名的 Codex 重置按钮供体验，欢迎来玩！

OpenClaw🦞@openclaw · 3小时前29

Thank you, @colinsolvely, for helping move the needed iOS and Android UI improvements forward with this post. We want to meet this feedback in the same spirit. Please share what is not working here or in Discord so the team can listen, respond, and turn that feedback into improvements.

译OpenClaw 项目官方发推感谢 @colinsolvely（Colin）推动 iOS 和 Android UI 改进，并鼓励用户继续在 X 或 Discord 反馈问题。Colin 在引用推文中自我介绍，称自己是负责 OpenClaw iOS UI 的志愿者维护者，时间有限但热衷移动端 AI 智能体，坦承 UI 有问题并接受批评，邀请大家在 Discord 对话，共同改进。双方强调社区共建。

Odyssey@odysseyml · 3小时前20

Probably staying a little longer.

译可能多待一会儿。

swyx @aiDotEngineer WF@swyx · 4小时前17

everyone come down to the AIE Expo at 12.30, there's a SPECIAL **FLASH** SURPRISE (you'll know it when u see it) and then come to Expo Stage 2 for a very special @latentspacepod live pod with Etched!

译大家12:30来AIE Expo，这里有特别的**闪现**惊喜（看到就知道），然后到Expo 2号舞台参加@latentspacepod与Etched的现场播客！

jason@jxnlco · 4小时前10

I will post a better version of the slides and share this with everyone.

译我将发布更高质量的幻灯片并与大家分享。

Artificial Analysis@ArtificialAnlys · 4小时前12

On Monday, Artificial Analysis brought together our largest SF community gathering yet, with conversations on model intelligence, evaluation, and AI performance, including our Intelligence Index v4.1 and our new long-horizon knowledge work benchmark, AA-Briefcase. Special thanks to Ivan Leo (@ivanleomk) @ Google Deepmind, Venkat Srinivasan @ NVIDIA, and Olive Song (@olive_jy_song ) @ MiniMax. Photo credit: @peterliuphotos

译周一，Artificial Analysis 举办了迄今为止规模最大的旧金山社区聚会，讨论了模型智能、评估和 AI 性能，包括我们的智能指数 v4.1 以及新的长时程知识工作基准 AA-Briefcase。 特别感谢 Google DeepMind 的 Ivan Leo (@ivanleomk)、NVIDIA 的 Venkat Srinivasan 和 MiniMax 的 Olive Song (@olive_jy_song )。 摄影：@peterliuphotos

🚨 AI News | TestingCatalog@testingcatalog · 5小时前58

Anthropic 🤝 Samsung Anthropic may develop its own AI chip with Samsung, according to The Information. > OpenAI recently announced their own AI chip developed in a partnership with Broadcom. Everyone is going full stack 👀

译Anthropic 🤝 Samsung 据The Information报道，Anthropic可能与三星合作开发自己的AI芯片。 > OpenAI近日宣布与Broadcom合作开发自有AI芯片。 大家都在走向全栈 👀

xAI@xai · 5小时前38

Grok Build is now installed in Railway sandboxes

译Grok Build 现已安装在 Railway 沙箱中

SemiAnalysis@SemiAnalysis_ · 5小时前28

EMIB-T Roadmap, Custom HBM, HBM4 Packaging Challenges, Microfluidic Cooling, Photonic Interconnects, and More ECTC 2026 Roundup, Intel, TSMC, SK Hynix, Samsung, Micron, Marvell, Lightmatter, Microsoft https://newsletter.semianalysis.com/p/ectc2026

译EMIB-T 路线图、定制 HBM、 HBM4 封装挑战、微流冷却、 光子互连，以及更多 ECTC 2026 综述，英特尔、台积电、SK 海力士、 三星、美光、Marvell、Lightmatter、微软 https://newsletter.semianalysis.com/p/ectc2026

Replit ⠕@Replit · 7小时前19

Check out what people are building on Replit: Open Thread 🧵

译来看看人们在 Replit 上构建了什么： 开放讨论串 🧵

Satya Nadella@satyanadella · 7小时前64

The future of the firm is a learning loop in which human capital and token capital compound. With our new Frontier Co., our ambition is to help every enterprise build its own AI capability, and to help create a frontier ecosystem where every organization can turn its knowledge, workflows, and judgment into its own AI systems that continuously improve. https://blogs.microsoft.com/blog/2026/07/02/microsoft-frontier-company-ai-engineering-that-amplifies-and-protects-your-intelligence/

译企业的未来是一个学习循环，人力资本与token资本不断累积。 借助我们的新Frontier Co.，我们的目标是帮助每个企业构建自己的AI能力，并帮助创建一个前沿生态，使每个组织都能将其知识、工作流程和判断力转化为自己的AI系统，并持续改进。https://blogs.microsoft.com/blog/2026/07/02/microsoft-frontier-company-ai-engineering-that-amplifies-and-protects-your-intelligence/

SiliconFlow@SiliconFlowAI · 8小时前31

Want to understand a repo without reading every file🔍 Point GLM-5.2 at any project you star🌟 It reads and builds you an interactive walkthrough All it takes: 🔑 a SiliconFlow API key ⚙️ plug GLM-5.2 into your agent 💬 point it at the repo, ask for a guide Now it's not just building — it's earning 🌊 SiliconFlow Summer Rush is LIVE: → Post what you build with GLM-5.2 on X → tag @SiliconFlowAI + #GLMOnSiliconFlow → Fill the form — you're in 🎉 More valid GLM-5.2 usage = higher leaderboard rank 📈 First 72h → Early Bird reward pool 🐦 📋 Register & full rules 👇 https://www.siliconflow.com/blog/siliconflow-summer-rush%E4%B8%A8glm-5.2-week-the-more-glm-5.2-you-run-the-more-you-win

译硅基流动举办Summer Rush-GLM 5.2 Week活动。使用GLM-5.2分析任意GitHub仓库，自动生成交互式代码解读向导。参赛需申请SiliconFlow API Key，将GLM-5.2接入智能体并指向仓库。6月29日20:30至7月6日20:30（PDT）期间，发布成果到X并标记@SiliconFlowAI和#GLMOnSiliconFlow，填写报名表即可参与排名。第1名获本周GLM-5.2消耗全额代金券返还（最高$1,000）及额外$50代金券。前72小时投稿享Early Bird额外奖励，所有有效参与均有抽奖机会。

Kling AI@Kling_ai · 8小时前25

For making it this far 🏆

译因为走到这一步 🏆

Chubby♨️@kimmonismus · 9小时前55

Huge:Anthropic is reportedly in early talks with Samsung to manufacture its own custom AI chip. The Information says the Claude maker is still figuring out what the processor should do, how powerful it should be, and how it would fit into servers or clusters. Anthropic already uses AWS Trainium, Google TPUs, and Nvidia GPUs, and says those will remain central. A custom chip would give it another lever as model deployment costs, data-center capacity, memory supply, and power all become strategic constraints. Samsung would get a high-profile AI foundry customer for its 2-nanometer process and advanced packaging push.

译The Information报道，Anthropic正与三星早期洽谈制造自研AI芯片，目前仍在规划芯片功能、算力及服务器/集群适配方案。Anthropic现有AWS Trainium、Google TPU、Nvidia GPU仍为核心，自研芯片将作为应对模型部署成本、数据中心容量、内存供应、电力等战略约束的新杠杆。三星借此获得高端AI代工客户，推动其2纳米制程和先进封装业务。

Chubby♨️@kimmonismus · 8小时前26

Since Fable 5 is expensive, smart routing acts as a moat.

译Claude Fable 5 已回归 AI 平台 ZenMux，并开放免费试用。ZenMux 同步推出 Auto Top-up Bonus 活动：充值可获 20% 额外积分，每人最多 3 次奖励，每次最高 $100，总计 $300。主推文指出，由于 Fable 5 价格昂贵，智能路由（smart routing）功能成为平台的护城河，能够有效降低成本。

Claude@claudeai · 9小时前29

Announcing Built with Claude: Life Sciences, a global virtual hackathon. Join us and @GladstoneInst for a week of researching and building with Claude Science and Claude Code, with a prize pool of $100k in credits.

译宣布「Built with Claude: Life Sciences」全球虚拟黑客马拉松。 与我们和@GladstoneInst一起，用Claude Science和Claude Code进行一周的研究与构建，奖金池为10万美元信用额度。

Kling AI@Kling_ai · 12小时前31

🏆 Cannes Lions Bronze Winner — Lorem Ipsum Samurai, cowboys, mafia, why are they all speaking “Lorem Ipsum”? Bronze Lion winner in Cannes Lions Film: B2B, Lorem Ipsum was created by Argentine studio Purga Films to promote its AI advertising business. Though every character speaks in meaningless placeholder words “Lorem Ipsum”, the emotions still hit hard. As the film says: “We have the craft. We need the scripts.” The entire film was created with Kling AI. Across wildly different styles and worlds, Kling delivered consistent performances, emotional depth, and cinematic control throughout the piece. Some stories still hit you deeply — even when you don’t understand a single word. Huge congrats to Purga Films on the win!

译可灵Kling AI官方宣布，由其生成的广告片《Lorem Ipsum》荣获戛纳狮子奖（Cannes Lions）电影类B2B铜奖。该片由阿根廷工作室Purga Films制作，片中所有角色均使用无意义的占位词“Lorem Ipsum”对话，但情感表达依然强烈。全片通过Kling AI生成，涵盖多种风格和世界，展现了可灵在表演一致性、情感深度和电影级控制上的能力。

小互@xiaohu · 13小时前32

兄弟们 福利来了 ChatGPT 促销，五折优惠 Plus会员只要10美金... 目前看只对Plus会员有折扣，其他会员无法享受 优惠链接在2楼↓

🚨 AI News | TestingCatalog@testingcatalog · 15小时前75

OPENAI 🔥: US government may get a 5% stake in OpenAI worth $42.6 billion, according to Financial Times and CNBC. > OpenAI proposes handing the U.S. government a 5% stake in the company, according to a report in the Financial Times. > The potential holding would be worth roughly $42.6 billion at the artificial intelligence startup’s recent $852 billion valuation. > OpenAI CEO Sam Altman reportedly argued the move was the best way to share the upside of AI with the public.

译据Financial Times和CNBC报道，OpenAI提议向美国政府提供公司5%的股份，按近期8520亿美元估值计算，价值约426亿美元。OpenAI CEO Sam Altman表示，此举是与公众分享AI发展红利的最佳方式。

Chubby♨️@kimmonismus · 15小时前67

OpenAI proposes handing Trump administration 5% stake. Heres why: According to FT, Sam Altman has discussed giving the US government a 5% stake in OpenAI as the $ 852bn startup faces rising scrutiny from Washington. Sams pitch: if AI creates enormous wealth, the public should own part of the upside. OpenAI has already proposed a “public wealth fund” that “provides every citizen - including those not invested in financial markets - with a stake in AI-driven economic growth.” The talks are early, may need Congress, and other AI labs have not agreed to mirror the idea. Imho the real reasons are the following: Regulatory insurance. A 5 percent stake would probably be cheaper than years of political gridlock, strict release rules, or a later special tax on AI profits. IPO preparation. OpenAI wants to go public eventually. Having a US state as an indirect stakeholder could make political risks appear smaller to investors. Model-release pressure. The article says OpenAI and Anthropic have delayed new frontier models due to US scrutiny. A public stake would be one way to turn Washington from an opponent into a co-beneficiary. Data centers, energy, permits. AI is no longer just software. OpenAI needs power, land, chips, local permits, and political support. A government stake could help with infrastructure expansion.

译据FT报道，OpenAI CEO Sam Altman与特朗普政府讨论给予美国政府5%股份，理由包括AI创造财富应让公众分享。背后原因：作为监管保险避免政治僵局；为IPO降低投资者政治风险感知；将华盛顿变为共同受益者以缓解前沿模型发布压力；换取数据中心、能源等基础设施支持。目前讨论尚处早期，需国会批准，其他AI实验室尚未同意跟进。

PixVerse@PixVerse_ · 17小时前37

@hq4ai Thank you for sharing this amazing workshop! It’s inspiring to see your amazing creation . Excited to support more creative education and future designers.

译日本大阪成蹊大学服装设计系使用 PixVerse 平台开设 AI 服装设计工作坊，学生在两个小时内完成从图像到视频的高质量闭环交付。PixVerse 官方感谢分享，并表示支持更多创意教育。

Alibaba Cloud@alibaba_cloud · 18小时前35

🥉 3rd Place at the AI Film Festival Monaco Hackathon! 🎬 Introducing 《The Glow of First Love》 by Iuliia Kiseleva — a hauntingly beautiful one-minute film that earned third place among global creators. After her husband dies in a 2003 car crash, a pregnant woman raises their daughter alone—but never truly alone. For 53 years, an unseen figure of light watches over her… until her final breath, when the glow fades and they reunite as they were at the beginning. A quiet, devastating meditation on enduring love and presence after loss—with light itself as a character—crafted using festival-grade AI tools that turn emotion into ethereal visual poetry. This is what’s possible when intimate storytelling meets Happy Horse.✨ Discover the creative engine behind it: https://int.alibabacloud.com/m/1000415018/

译阿里云Happy Horse平台创作的短片《The Glow of First Love》在摩纳哥AI电影节黑客松中获得第三名。该一分钟作品讲述一名女子在2003年丈夫车祸去世后独自抚养女儿，53年间被无形光之守护者陪伴，直至临终重逢。短片利用Happy Horse平台将情感转化为诗意视觉，展示了festival-grade AI工具在叙事中的潜力。

宝玉@dotey · 18小时前60

OpenAI 提议向美国政府出让 5% 股份：让普通人也能共享“AI 红利” OpenAI 正在酝酿一项史无前例的计划：这家估值高达 8520 亿美元的人工智能初创公司，正探讨将 5% 的股份交给美国政府。 据知情人士透露，自从特朗普总统开启第二任期以来，OpenAI 首席执行官山姆·奥特曼（Sam Altman）一直在与多位美国政府高官进行初步讨论，探讨联邦政府入股大型人工智能公司的可能性。早在 2025 年初，奥特曼就直接向特朗普总统提出了这个构想，希望通过这种让公众在公司中拥有经济利益的方式，来分享 AI 带来的好处，同时也借此扫清近期的政治障碍。 为什么要采取如此罕见的举措？因为人工智能的发展速度已经令人震撼。那些不久前还只存在于科幻小说里的系统，现在已经被全球各地的企业和政府广泛部署。AI 在经济价值、国家安全以及加速科学发现方面的重要性已经非常清晰。预计只需再过一两年，人类就能打造出威力惊人的系统，为世界带来巨大价值。这项技术对人类物质生活条件的重塑，规模将堪比甚至超越电力的利用。 为了应对这种足以改变世界的财富大爆炸，相关提案提出了建立“公共财富基金”（Public Wealth Fund）的构想。 这个基金能给普通人带来什么实际好处？简单来说，它就像是一个全民共享的分红池。政策制定者和 AI 公司将合作提供初始资金，投资于具有长期增长潜力的多元化资产——既包括 AI 公司本身，也包括更广泛的采用和部署 AI 技术的其他企业。基金的收益将直接发放给每一位公民。这意味着，不管你最初的财富水平如何，不管你平时有没有途径投资金融市场，你都能直接参与并享受到由 AI 驱动的经济增长所带来的红利。

译OpenAI 正探讨将 5% 股份交给美国政府，建立“公共财富基金”。估值 8520 亿美元的 AI 初创公司 CEO 山姆·奥特曼自 2025 年初向特朗普总统提出构想，旨在让公众通过持股分享 AI 经济收益，同时扫清政治障碍。基金将投资 AI 公司及采用 AI 技术的企业，收益直接发放给公民，使普通人无需参与金融市场也能享受 AI 驱动增长的红利。

郭明錤｜Ming-Chi Kuo@mingchikuo · 18小时前60

這個結構性的轉變，代表 Amazon 做好了為 AI 算力擴張的長期準備。意味著 Amazon 不認為有 AI 算力過剩的問題。

译郭明錤产业调查显示，Amazon自有消费电子（Kindle、Fire TV、Echo等）首次放弃外购处理器，改采与AI芯片Trainium类似的COT模式，由世芯-KY独家负责后段设计与测试。受AI算力扩张影响，截至1Q26的12个月自由现金流年减95%至约12亿美元，Amazon借此优化非AI业务成本以支撑长期AI投资。策略预计2027年启动，全面转换后自研处理器年出货约4000万颗，表明Amazon不认为AI算力过剩。

郭明錤｜Ming-Chi Kuo@mingchikuo · 18小时前55

This structural shift suggests that Amazon has been preparing for the long-term expansion of AI compute, and that it does not see AI compute oversupply as a real issue.

译郭明錤调研显示，Amazon消费电子（Kindle、Fire TV、Echo等）将首次改变20年来外购处理器策略，转向COT自研模式，由Alchip独家负责后端设计与测试。背景是AI算力快速扩张导致截至2026年Q1的12个月自由现金流下降95%至约12亿美元，Amazon需优化非AI业务成本以维持AI投资。新策略预计2027年启动，完成后年自研处理器出货约4000万。主推文认为，Amazon正为AI算力长期扩张做准备，并不认为AI算力过剩是真实问题。

Alibaba Cloud@alibaba_cloud · 18小时前19

Zero downtime and zero data loss migration? Make it a reality by migrating to Alibaba Cloud. Agrinesia streamlined operations, reduced infrastructure spend, and got future-ready with help from AI-powered Qwen. Read more here: https://int.alibabacloud.com/m/1000415075/

译零停机零数据丢失的迁移？迁移到阿里云即可实现。Agrinesia借助AI驱动的Qwen，简化了运营、降低了基础设施支出，并为未来做好准备。 了解更多：https://int.alibabacloud.com/m/1000415075/

DogeDesigner@cb_doge · 19小时前55

BREAKING: OpenAI has reportedly discussed giving the U.S. government a 5% ownership stake. Sam Altman says this would help ensure Americans share in the success of AI. He has also proposed that other leading U.S. AI companies consider doing the same, as per The Financial Times.

译据报，OpenAI 已讨论给予美国政府 5% 的所有权股份。 Sam Altman 表示，这将有助于确保美国人分享 AI 的成功。他还提议其他领先的美国 AI 公司考虑同样的做法，据《金融时报》报道。

Rohan Paul@rohanpaul_ai · 19小时前64

Godot is banning vibe coding after AI-made PRs turned review time into the bottleneck. Substantial AI-generated code will also be barred, while small aids like completion remain allowed. It's an open-source game engine, so outside contributors constantly send proposed code changes. Every PR still needs a maintainer who understands the engine deeply enough to spot risk. AI changed the cost balance because generating code got cheaper, but reviewing code stayed expensive. Reviewer capacity was already too thin, and AI-made submissions made it harder. Contributors must disclose any AI help used while writing code for a PR. Godot is also banning AI-generated text in PR discussions, issues, and proposals. imo, enforcement will be impractical, they probably will never know with certainty what was vibe coded and what was not, and that is the whole weakness of the rule. --- godotengine .org/article/contribution-policy-2026/

译Godot 开源游戏引擎发布新贡献政策，禁止“vibe coding”（AI 生成大量代码），因 AI 生成的 PR 使审阅时间成为瓶颈。大量 AI 生成代码将被禁止，仅允许代码补全等小型辅助工具。贡献者需披露是否使用 AI 辅助编写代码，同时禁止在 PR 讨论、议题、提案中使用 AI 生成文本。推文作者认为该规则执行不切实际，难以区分哪些代码是 AI 生成的。

Rohan Paul@rohanpaul_ai · 18小时前59

Palo Alto Networks CEO Nikesh Arora said 90% of enterprise workers are behind on AI, and it could determine the fate of their careers. He expects 20% to 25% of his workforce to change within 12 months. According to a 2025 Orgvue study, 39% of leaders had already cut workers after deploying AI. --- fortune .com/2026/07/01/ceo-of-palo-alto-networks-nikesh-arora-workers-about-to-face-darwinian-moment-thanks-to-ai-evolve-or-get-cut/

译Palo Alto Networks CEO Nikesh Arora称，90%的企业员工在AI方面已经落后，而这可能决定他们的职业命运。 他预计其员工队伍中20%至25%的人将在12个月内发生变化。 根据2025年Orgvue的一项研究，39%的领导者已在部署AI后裁员。

郭明錤｜Ming-Chi Kuo@mingchikuo · 19小时前49

My latest industry checks indicate that Amazon’s processor procurement strategy for its own consumer electronics is set to undergo its first major shift in 20 years: moving away from externally sourced processors and adopting a COT (customer-owned tooling) model, with Alchip as the exclusive provider of back-end design and testing for its self-developed chips. Key points: 1. Amid the rapid expansion of AI compute, Amazon’s free cash flow for the 12 months ended 1Q26 fell 95% year over year to about US$1.2 billion. To maintain financial flexibility and keep funding its AI investment cycle, Amazon is also streamlining its organization and improving the cost structure of its non-AI businesses. 2. Amazon's own-brand consumer electronics, including Kindle, Fire TV, Echo, Alexa-enabled products, Blink, and Ring, currently use externally sourced processors. To optimize its cost structure, Amazon plans to gradually shift away from external sourcing and adopt a COT model similar to the one used for AI chip Trainium, taking more control of processor development in-house. 3. For this transition, Amazon has selected Alchip as its exclusive partner for back-end design and testing. Alchip is expected to receive non-recurring engineering, or NRE, fees for each design project, while also benefiting from processor shipments. 4. The strategy is expected to begin in 2027. Once the transition is complete, annual shipments of Amazon’s in-house processors are estimated at around 40 million units. This shift should help improve the cost structure of Amazon’s own devices and become a meaningful growth driver for Alchip.

译郭明錤产业调查显示，Amazon拟对其消费电子处理器采购进行20年来首次重大调整。因AI计算扩张，截至1Q26的12个月自由现金流同比降95%至约12亿美元，为优化非AI业务成本，Amazon将逐步放弃Kindle、Fire TV、Echo等自营设备的外部处理器，采用类似Trainium的COT模式，将处理器开发内包。已选定Alchip为独家后端设计与测试合作伙伴，收取NRE费用并受益于芯片出货。策略预计2027年启动，完成后年出货约4000万颗，有望改善设备成本并成为Alchip增长动力。

Baidu Inc.@Baidu_Inc · 19小时前33

Glad you enjoyed the ride🤩

译在迪拜朱美拉区乘坐百度Apollo Go。 全区域无人驾驶服务仅需5迪拉姆，约1.36美元。 没有安全员，位于主要旅游区，比一杯咖啡还便宜。 很疯狂。🤯 百度官方回应："Glad you enjoyed the ride🤩"

Alibaba Cloud@alibaba_cloud · 20小时前26

Wrapped up our 1st PH AI Scenario Workshop on June 18! 🚀 Welcomed 80+ engineers for deep dives into Alibaba Cloud's AI-native stack that transforms enterprise operations at scale. Huge thanks to our speakers & attendees. Looking forward to helping more Philippine enterprises build intelligent, self-driving infrastructure! #AlibabaCloud #AIOps

译我们在6月18日结束了首次PH AI场景工作坊！🚀 欢迎了80多位工程师，深入探讨阿里云AI原生技术栈，该技术栈可在规模上转变企业运营。感谢我们的演讲者和与会者。期待帮助更多菲律宾企业构建智能、自驱的基础设施！#阿里云 #AIOps

MiniMax (official)@MiniMax_AI · 22小时前20

Scale up in Every Format. @aiDotEngineer #MiniMax

译在每个格式中扩展规模。 @aiDotEngineer #MiniMax

小互@xiaohu · 22小时前43

Fable 5 只有一周使用时间 过后就只能通过API 来使用，无法使用订阅的用量 另外Claude 今天重置了所有人的用量 可劲造吧

ginobefun@hongming731 · 1天前43

BestBlogs 早报 · 07-02 # LongCat-2.0 / 本地 AI / 美图 AI 方法论 / Google ADK 2.0 / Claude Fable 5 [1] ★ 精讲｜Ahmad Osman 谈本地 AI 为何正在追赶 http://Latent.Space 专访 Osmantic 创始人 Ahmad Osman，这位长期倡导本地 AI 的开发者在 AIEWF 办了两场爆满 workshop。他判断开源与闭源前沿模型的差距正持续缩小，目前大约落后 4 到 8 个月。他用朋友买 RTX 5090 跑 Qwen 3.5 改 RGB 灯光失败的例子说明：本地 AI 缺的不是模型，而是搜索、工具、Agent 等完整栈。他从企业主权算力和混合架构角度，讲清了为什么本地 AI 正被认真当作基础设施。 来源：http://Latent.Space https://www.bestblogs.dev/article/a6371e93 [2] ★ 精讲｜美团 LongCat-2.0 正式发布：在国产算力集群上完成全流程训练与推理的万亿参数模型 美团技术团队官方披露 LongCat-2.0 万亿参数 MoE 模型（总参数 1.6T，平均激活约 48B），在 5 万卡国产算力集群上完成全流程训练与推理。预训练数据超 30T tokens，月均日故障率降低 70% 以上，训练 MFU 提升 1.5 倍。SWE-bench Pro 得分 59.5，超过 GPT-5.5 与 Claude Opus 4.6。原生支持 1M 上下文，已跻身 OpenRouter 全球调用量前三。值得读在于，它是少数把国产算力、万亿 MoE、Agentic Coding 全链路讲透的工程实录。 来源：美团 · 技术团队 https://www.bestblogs.dev/article/ad5a0b93 [3] ★ 精讲｜专访美图 CEO 吴欣鸿：做 AI 产品，是一场难以提前策划的游戏 《智能涌现》专访美图 CEO 吴欣鸿。美图 2025 年营收 38.58 亿元、净利润 9.65 亿元（同比 +64.7%），AI 重构的影像设计收入占比从 35% 升至 76.6%。他立规矩：新产品立项到上线不超 1 个月，半年 ARR 须达 10 万美元，且老产品禁止导流。MVLAND 内测两三个月 ARR 就到 10 万美元、现已近 50 万美元。值得读在于，他用自然生长而非策划、热爱而非纯 PMF 的方法论，讲清了一家 2000 人公司如何在 AI 应用层持续跑赢。 来源：智能涌现 https://www.bestblogs.dev/article/8b6cc4f7 [4] 为什么我们构建了 ADK 2.0 Google ADK 2.0 引入了一个结构化的工作流运行时，将确定性代码执行与 LLM 智能体相结合，解决了生产环境中的可靠性问题，如循环、幻觉和高成本。 来源：Google Developers Blog https://www.bestblogs.dev/article/76d5c422 [5] RAG 的上下文工程：每个 RAG 答案背后的四种类型化输入 本文通过上下文工程的视角重新审视单文档 RAG，展示了如何让管道中的每个组件输出类型化输入，并汇聚成一次可审计且成本高效的 LLM 调用。 来源：Towards Data Science https://www.bestblogs.dev/article/33fa6204 [6] 高德 GrowLoop：构建感性对话的理性 Benchmark 高德团队提出 GrowLoop 系统，通过启发式学习与双循环协进化机制，将感性对话评判标准转化为可生长的理性 Benchmark，解决开放域对话真人感评测难题。 来源：AI 前线 https://www.bestblogs.dev/article/4bedb1a9 [7] 我们团队从 AWS 迁移到 PaaS 的经历 一个 7 人团队量化了 AWS 基础设施维护的隐性成本，在 3 周内迁移到 Sevalla PaaS，并每周节省 10 小时工程师时间。 来源：freeCodeCamp https://www.bestblogs.dev/article/49006840 [8] 人类-AI 交互设计的 39 条原则 一个包含 39 条人类-AI 交互设计原则的综合框架，按九个主题组织：概率基础、预期设定、校准信任、透明度、控制、优雅失败、共同创造、负责任自主和持续依赖。 来源：UX Collective https://www.bestblogs.dev/article/2cba6a6e [9] 如何把超级个体的产能，转化成组织能力？ | AI 跃迁者调研 本文深度访谈出门问问 CEO 李志飞，揭示从超级个体到超级组织的转型路径：以自研 CodeBanana 系统实现沟通与执行合一，通过全栈转型与系统设计师机制将 AI 产能转化为组织能力。 来源：腾讯研究院 https://www.bestblogs.dev/article/e3a487f9 [10] AI UITester：AI Native 的 UI 自动化测试新范式｜得物技术 本文介绍得物技术团队自研的 AI 原生 UI 测试工具 ai_uitester，通过 VLM 视觉驱动、LLM 用例自动生成和 AI 自愈调试，实现三端统一运维，大幅降低测试维护成本。 来源：得物技术 https://www.bestblogs.dev/article/694f9d01 --- http://BestBlogs.dev · 发现真正适合你的高质量内容 BestBlogs 是 AI 驱动的私人阅读助手，帮助你发现真正适合你的高质量内容，欢迎体验。 在线阅读：https://www.bestblogs.dev/explore/brief/2026-07-02

译美团发布LongCat-2.0万亿MoE模型（总参1.6T、激活48B），在5万卡国产算力集群完成全流程训练，SWE-bench Pro得分59.5超GPT-5.5与Claude Opus 4.6，原生支持1M上下文，已跻身OpenRouter全球调用量前三。本地AI倡导者Ahmad Osman称开源与闭源前沿差距缩小至4-8个月，但缺少搜索、工具等完整栈。美图2025年营收38.58亿元、净利润9.65亿元（同比+64.7%），AI收入占比升至76.6%，新产品上线不超1个月、半年ARR达10万美元。Google ADK 2.0引入结构化工作流运行时，结合代码执行与LLM智能体解决可靠性问题。

Replit ⠕@Replit · 1天前26

Some moments from vibecon that stuck with us. A few days of big ideas, fast builds, and people who show up ready to make something real. Thank you to everyone who joined, watched, and shared what they're working on. This is what building with Replit looks like. 🚀

译一些在 vibecon 上让我们难忘的瞬间。 几天时间里，大胆的想法、快速的构建，以及那些准备好创造真实事物的人们。感谢每一位参与、观看和分享他们正在制作的内容的人。 这就是用 Replit 构建的样子。🚀