关注最新看重点大公司财经动态看世界硅谷动态金融圈金融动态追踪现货黄金盘中涨破4400美元。8月11日早盘，现货黄金持续拉涨，盘中突破4400美元/盎司，涨超0.2%，创自6月8日以来新高。过去一周，现货黄金涨超7%，创今年1月19日以来最大单周涨幅。当地时间8月10日，现货黄金延续上涨态势，盘中最高触及4395美元/盎司。11秒前00举报发表0/500虎嗅用户社区交流公约IPO动态又有哪家公司上市了？今日3只新股上市，涉及新材料细分领域“小巨人”等。今日无新股申购；②有3只新股上市，分别为创业板的超纯应材，科创板的国仪公司，北交所的珈凯生物。1分钟前00举报发表0/500虎嗅用户社区交流公约莫言财经财经资讯，知名财经评论信息收集中金：AI硬件降本多元化之路开启后有望刺激更多需求。中金公司研报称，当下，在存储器成本持续通胀、海外CSP自由现金流开始出现转负的背景下，市场焦点再次转回“AI硬件投资回报率”这一核心问题，7月AI硬件赛道股大幅调整。与上一轮下跌不同，此次大厂资本开支、AI Lab年化收入等核心指标已处在高位，市场交易拥挤度也高于前期。通过对单token（词元）成本、AI Lab API利润等核心问题研究结论，中金公司认为AI硬件需求增长依然强劲。硬件成本端，中金公司认为存储通胀难抵GPU技术迭代增效，单token（词元）成本继续快速下降。模型端来看，闭源模型靠溢价、开源模型凭降本，各厂商盈利底座稳固，上游算力硬件需求持续被拉动。另外，推理负载结构的精细化有望推动定制化架构芯片落地及系统互联创新。随着推理应用从传统Chat-only向Agent等转换，AI硬件或通过定制化架构，系统互连优化加速单token（词元）降本、或突破延迟与吞吐瓶颈，为实时交互和Agent场景提供性能溢价。硬件降本多元化之路开启后有望刺激更多需求。1分钟前00举报发表0/500虎嗅用户社区交流公约飞行圈航空领域靠谱资讯速递~国产大飞机C919将于明日起首次执行国际商业航班。据中国国航消息，8月12日起，国航北京首都—乌兰巴托航线CA723/4航班，将启用国产飞机C919执飞，这是C919首次执行国际商业航班。5分钟前00举报发表0/500虎嗅用户社区交流公约金融圈金融动态追踪机构：果断的买盘压力铸就黄金涨势。Pepperstone研究主管Chris Weston表示，黄金已重新牢牢回到交易员的视线中。他说，在突破了从6月下旬一直维持到8月3日的盘整区间后，黄金和白银都出现了上涨，目前金价正在测试每盎司4,400美元。Weston表示，突破7月高点伴随着果断的买盘压力，随后的势头表明多头占据了主导地位。客户活动也有所增加，头寸偏向多头。他补充说，目前所有未平仓黄金头寸中有60%是多头。5分钟前00举报发表0/500虎嗅用户社区交流公约安全小喇叭重大灾害事故预警、播报、跟进宁夏沙坡头通报“女孩体验高空滑索被卡半空”：因风向突变致游客在空中滞留1分钟，已安抚道歉，将根据实时气象条件动态调整项目运营安排。8月11日，中旅（宁夏）沙坡头索道游乐有限公司发布关于沙坡头飞黄腾达单绳滑索游客高空滞留情况的说明：8月7日15时52分，一名女性游客在沙坡头景区体验“飞黄腾达”单绳滑索项目时，在滑行过程中因风向突变，导致游客在空中滞留约1分钟。经现场工作人员手动调整钢丝绳，引导游客调整滑行姿态，滑行小车顺利带动游客安全滑行进站。游客抵达终点后，工作人员立即对其进行情绪安抚和解释道歉，游客对此次因风向突变导致短暂滞留表示理解。下一步，我司将根据实时气象条件动态调整项目运营安排，同时加强黄河滑索项目操作人员应急处置培训，确保广大游客人身安全。此前报道：近日，宁夏中卫，一名女孩在体验滑索项目时，滑到半路突然停住，被困在半空中，经自行挣扎后脱困，顺利滑至终点。景区工作人员回应称，滑索项目对参与者有明确的体重限制，必须达到70斤以上才可体验。此外，风力等自然天气因素也可能影响滑索的正常滑行状态。13分钟前00举报发表0/500虎嗅用户社区交流公约一只飞虎最近国外发生了什么？朝中社：日本把朝鲜等周边国家描写为“重大而紧迫的威胁”，企图使其扩军备武行为合理化，“侵略基因”一直存在于日本的政治风土中。据朝中社11日报道，朝鲜国际安保问题评论家发表文章称，国际社会谴责日本2026年版《防卫白皮书》是再侵略白皮书。文章说，日本把否定历史犯罪和实现军事大国化作为“国策”，比以往任何时候都进一步强化同美国等势力的同盟。日本在《防卫白皮书》中把朝鲜等周边国家描写为“重大而紧迫的威胁”“最大的战略挑战”，企图使其扩军备武行为合理化。文章认为，“侵略基因”一直存在于日本的政治风土中，且越来越危险。日本企图披着“防卫”外衣，掩盖其侵略本质，但无法掩盖其蹂躏地区和平、加剧紧张局势的祸根实质。文章呼吁，国际社会应对严重危害地区和平与安全、日趋危险的日本新军国主义活动提高警惕。21分钟前00举报发表0/500虎嗅用户社区交流公约安全小喇叭重大灾害事故预警、播报、跟进哥伦比亚强震已致111人死亡、87人受伤，哥政府宣布进入“国家灾难状态”，强震瞬间画面显示，地面晃动、建筑倒塌、民众奔逃。据央视新闻：哥伦比亚西部10日上午发生7.5级地震，全国多地震感强烈。现场视频记录下地震发生的瞬间，地面剧烈晃动，多座建筑受损倒塌，现场烟雾四起，有民众在避难时摔倒。哥伦比亚总统德拉埃斯普列亚10日在首都波哥大通报说，该国西部当天发生的强震已造成全国111人死亡、87人受伤。哥伦比亚政府已宣布进入“国家灾难状态”，应对地震灾情。23分钟前00举报发表0/500虎嗅用户社区交流公约金融圈金融动态追踪破解小币种跨境结算繁难，银行组合拳赋能外贸企业出海。记者走访多地外贸企业与商业银行发现，银行业正打出跨境金融服务组合拳。今年以来一些银行纷纷上线小币种直兑清算业务，省去美元中转环节，压降企业汇兑成本，部分银行结合企业需求，差异化定制汇率避险方案。多方赋能下，跨境金融的“毛细血管”正被重新疏通。35分钟前00举报发表0/500虎嗅用户社区交流公约老·司机关心智能出行的汽车产业观察者中国汽车流通协会：7月汽车经销商库存系数为1.48，环比下降6.3%。8月10日，中国汽车流通协会发布2026年7月“汽车经销商库存”调查结果：7月汽车经销商综合库存系数为1.48，环比下降6.3%，同比上升9.6%。据中国汽车流通协会乘联分会发布的车市扫描数据显示，7月全国乘用车市场零售150.6万辆，以此测算，7月末汽车经销商库存总量约220万辆左右，渠道整体库存规模与6月末相比有所降低。37分钟前00举报发表0/500虎嗅用户社区交流公约科学咖搞科研的那帮人……我国高能量密度一次电池功率获突破。据科技日报，从中国科学院大连化学物理研究所获悉，该所能源催化转化全国重点实验室陈忠伟团队成功研制出25安时级软包金属锂一次电池。经中国电子技术标准化研究院安全实验室检测，该电池低倍率下能量密度超过750Wh/kg和1285Wh/L，属于一次电池中的高能量密度区间，放电倍率可提高百倍，这一成果标志着我国在高功率一次电池领域迈入国际前列，为高端装备装上自主“电力心脏”。42分钟前00举报发表0/500虎嗅用户社区交流公约读数数字里的故事港珠澳大桥珠海公路口岸进出境车辆累计突破2500万辆次。截至昨天（9日），港珠澳大桥珠海公路口岸进出境车辆已累计突破2500万辆次，单日进出境车辆最高峰达2.97万辆次，客车通关量居全国首位。今年以来，口岸进出境车辆已达425万辆次，其中港澳单牌车248万辆次，同比增长8.1%。43分钟前00举报发表0/500虎嗅用户社区交流公约AI新技能机器人们又get了什么新本领？私募巨头集体入局，据称英伟达将牵头5000亿美元融资大单。知情人士透露，多家金融巨头将与英伟达合作，筹集5000亿美元用于AI基础设施建设，或成华尔街迄今最大规模贷款行动之一；②私募资本正将数万亿美元投向AI基础设施。先前，摩根士丹利预计2026-2028年超大规模云服务商将投入约3.5万亿美元用于相关领域。45分钟前00举报发表0/500虎嗅用户社区交流公约货币的战争关于“钱”美联储鹰派释放强硬信号：一次升息不够，或需多次加息。哈马克称美联储或需多次加息，才能推动通胀率降至2%的目标水平，当前利率未对经济形成实质性限制；②两周前美联储维持利率不变，哈马克等三位票委反对；芝商所数据显示9月该行加息25基点概率过半。46分钟前00举报发表0/500虎嗅用户社区交流公约字节小报字节跳动对抗互联网老巨头豆包刘星：“豆包推荐酒店要收费”说法不准确，还处在早期探索。8月10日晚，豆包公关负责人刘星发文表示，“豆包推荐酒店要收费了”这个说法不准确。豆包生活服务业务目前没有付费推广，豆包推荐酒店也不收取广告费，商家不能通过付费影响推荐或排序，仅在订单成交后支付渠道服务费。部分平台采用“佣金+广告”的收费模式，豆包现阶段主要按成交收费，两者不好直接比较。对商家而言，更有参考意义的是实际渠道成本和新增订单效果，豆包也还在非常早期的探索，欢迎大家体验和提意见。51分钟前00举报发表0/500虎嗅用户社区交流公约新政新规各种新政策美股光通信大幅回落，中概股逆势走强。《中国人民银行“十五五”改革发展规划》发布②A股又一波回购潮来了③美股光通信股跳水、存储股分化④韩国将设立超35亿美元基金，重点投资半导体材料、零部件及设备领域⑤存储龙头上半年净赚超百亿。58分钟前00举报发表0/500虎嗅用户社区交流公约AI新技能机器人们又get了什么新本领？OpenAI据悉回购70亿美元员工股份，估值维持在8520亿美元。据报道，OpenAI已完成一项员工股份出售交易，规模约70亿美元。知情人士表示，OpenAI此次未引入外部投资者，而是自行回购现任和前任员工持有的股份。据悉，这笔交易对OpenAI的估值为8520亿美元，与最近一轮融资时估值持平。58分钟前00举报发表0/500虎嗅用户社区交流公约安全小喇叭重大灾害事故预警、播报、跟进中央气象台8月11日06时继续发布台风蓝色预警。今年第13号台风“白海豚”（热带风暴级）的中心今天（11日）早晨5点钟位于安徽安庆市境内，就是北纬30.2度、东经116.2度，外围最大风力有8级（20米/秒），中心最低气压为992百帕。预计，“白海豚”将以每小时15-20公里的速度向西偏北方向移动，强度逐渐减弱。59分钟前10举报发表0/500虎嗅用户社区交流公约读数数字里的故事受世界杯提振，英国零售额小幅增长。一份报告显示，随着英格兰队挺进国际足联(FIFA)世界杯后期阶段提振了食品和饮料销售，英国7月份零售额小幅增长。然而，根据英国零售商协会和毕马威的这份报告，由于气温飙升导致客流量减少，非食品类销售受到打击。报告称，在7月5日至8月1日期间，英国总零售额同比增长1.3%。“在高温下，消费者需求举步维艰，导致零售商在下半年伊始面临挑战，“英国零售商协会首席执行官Helen Dickinson说。Dickinson补充说，尽管7月份小幅增长，但由于充满挑战的经济环境，消费者的预算仍然紧张，同时零售商继续努力应对更高的成本。1小时前00举报发表0/500虎嗅用户社区交流公约股市短评这里有你关注的股势和异动美股纳指收跌0.3%，光通信板块大跌，Coherent跌超14%。美股三大指数集体收跌，道指跌0.13%，纳指跌0.32%，标普500指数跌0.06%。光通信板块高开低走，Coherent跌超14%，Lumentum跌超8%。存储板块涨跌不一，闪迪涨超2%，SK海力士、希捷科技跌超1%。“科技七巨头”多数上涨，亚马逊涨1.30%，微软涨1.23%，特斯拉涨0.70%，谷歌涨0.69%，Meta涨0.48%，苹果跌1.56%，英伟达跌2.87%。标普500能源指数上涨4.6%，创2025年4月以来最佳单日表现。1小时前00举报发表0/500虎嗅用户社区交流公约

ENTrendingAI Office ToolsAI Video ToolsAI Programming ToolsAI Chat AssistantAI Writing ToolsAI Learning WebsitesAI Image ToolsAI Training ModelsAI Search EngineAI Development PlatformAI Audio ToolsAI Design ToolsAI Model EvaluationAI Prompt InstructionsAI Content DetectionAI AgentsTrusted by 10M+ Users Worldwide - AI Tools HubAI Tutorials · Get Started FasterView MoreAI Pet Sword and Shield Short Video Trending Profit Case: Two Episodes with Over Ten Million Views, Creating a Comforting Virtual IPReal-Life Reverse COS AI Fruit - Trending Hot Topic, 35 Million Views in Two DaysAI-Generated Cutesy IP GuGuGaGa: The Full Process from Traffic Hit to Physical Product MonetizationBuilding a SaaS Packaging Tool with OpenClaw to Achieve a Compound Business with $200 Daily EarningsAI Heretic Composition Method: Use AI Sample Clips to Guide Real Photography, Easily Gain 30,000 Likes on XiaohongshuAI Eastern Monk: Building a Mindful Mentor Persona on Overseas Social Media Platforms - A Traffic Monetization Strategy to Attract Millions of FollowersTrendingView MoreDeepSeekDeepSeek was established in 2023 and focuses on researching the world’s leading general artificial intelligence underlying models and technologies. Based on self - developed training frameworks, self - built intelligent computing clusters, and computing power of tens of thousands of GPUs, the team released and open - sourced multiple large models with tens of billions of parameters in just half a year, such as the DeepSeek - LLM general large - language model and the DeepSeek - Coder code large model. In January 2024, it also became the first to open - source the first domestic MoE large model (DeepSeek - MoE). Each of these large models has excellent performance in public evaluation lists and generalization effects beyond real - world samples, outperforming models of the same level. Currently, it offers an API open platform, featuring good cost - effectiveness and openness. The product is positioned to provide users with high - performance and user - friendly artificial intelligence models, helping various industries make breakthroughs in their intelligent transformation.DeepSeekDeepSeek was established in 2023 and focuses on researching the world’s leading general artificial intelligence underlying models and technologies. Based on self - developed training frameworks, self - built intelligent computing clusters, and computing power of tens of thousands of GPUs, the team released and open - sourced multiple large models with tens of billions of parameters in just half a year, such as the DeepSeek - LLM general large - language model and the DeepSeek - Coder code large model. In January 2024, it also became the first to open - source the first domestic MoE large model (DeepSeek - MoE). Each of these large models has excellent performance in public evaluation lists and generalization effects beyond real - world samples, outperforming models of the same level. Currently, it offers an API open platform, featuring good cost - effectiveness and openness. The product is positioned to provide users with high - performance and user - friendly artificial intelligence models, helping various industries make breakthroughs in their intelligent transformation.Tencent YuanbaoTencent Yuanbao is an AI application developed based on Tencent’s Hunyuan large model, featuring a high degree of intelligence and multi - functionality. Its significance lies in the fact that it integrates various capabilities such as writing, painting, and searching, which can greatly improve users’ work efficiency and creative inspiration. The main advantages include comprehensive functions that can meet the needs of different scenarios; being intelligent and efficient, providing accurate and rapid answers and creative solutions. The product is backed by Tencent’s strong technological R & D capabilities. Currently, the document does not mention price information, and its positioning is to become an intelligent assistant for users to work easily and enrich their lives.Tencent YuanbaoTencent Yuanbao is an AI application developed based on Tencent’s Hunyuan large model, featuring a high degree of intelligence and multi - functionality. Its significance lies in the fact that it integrates various capabilities such as writing, painting, and searching, which can greatly improve users’ work efficiency and creative inspiration. The main advantages include comprehensive functions that can meet the needs of different scenarios; being intelligent and efficient, providing accurate and rapid answers and creative solutions. The product is backed by Tencent’s strong technological R & D capabilities. Currently, the document does not mention price information, and its positioning is to become an intelligent assistant for users to work easily and enrich their lives.CursorCursor is an AI code editor developed by Anysphere. It is powered by a combination of customized and cutting - edge models, which is fast an

ENDetect GEO Promotional Link Optimization EffectsQuickly assess how your promotional articles are referenced on AI platforms like DeepSeek, Doubao, Tongyi Qianwen, etc., using data to support decision-making.Step 1: Input Article Title0 / 80Step 2: Input Promotional LinkStep 3: Configure Detection Questions(Affects final score coverage)Smart Match Detection QuestionsThe system will automatically generate 5 core questions based on the article content for detectionCustom Detection QuestionsManually configure specific detection directions (requires 3 questions)Step 4: Select Detection PlatformStart AI Citation DetectionWhy Choose GEO Detection?Sent promotional materials, how do you know if they are effective? We help you evaluate how your content is cited on AI platforms.Quick AssessmentAbout 2 minutes after pasting the link, query China’s top 5 AI platforms to quickly obtain citation status.Exclusive GEO ScoringComprehensive assessment from three dimensions: ‘inclusion status’, ‘citation frequency’, and ‘question coverage’ to help you judge the AI exposure effect of promotional content.Citation TracingNot only see the score, but also see which platform mentioned your content when answering what question.Coming SoonBatch DetectionSubmit up to 100 links at once, process them uniformly in the background, and view them uniformly after completion.Effect TrackingHistorical records are automatically saved, supporting comparison at different time points to observe changes in AI exposure trends.Coming SoonReport ExportGenerate visual reports with citation details for easy reporting to clients or leaders.Complete Detection in Three StepsNo complicated configuration required, just paste the link to startStep 1: Paste LinkEnter the title of your published article, then paste the article URL. Supports single submission and intelligently extracts the core content.Step 2: Smart QueryThe system automatically queries relevant questions to various AI platforms, covering China’s top 5 mainstream AI platforms, parallel query to improve efficiency, real-time display of progress.Step 3: View ReportGet scores and citation details to understand how AI platforms cite your content, visual score display, detailed citation details, support export and sharing.SEO vs GEO: New Era Traffic Entry PointsTraditional SEO focuses on search engine rankings, while GEO focuses on whether your content is recommended by AI assistants to users.Comparison DimensionTraditional SEOGEO Generative OptimizationOptimization TargetGoogle/Baidu Search RankingsAI Platform Citations & RecommendationsKey MetricsKeyword Rankings, Click-through RatesAI Citation Rate, Answer CoverageUser BehaviorActive Search ClicksPassive AI Answer ReceptionImpact ScopeSearch Results PageAI Conversations, Smart AssistantsFuture TrendsGrowth StabilizingExponential GrowthOptimization TargetTraditional SEOGoogle/Baidu Search RankingsGEO Generative OptimizationAI Platform Citations & RecommendationsKey MetricsTraditional SEOKeyword Rankings, Click-through RatesGEO Generative OptimizationAI Citation Rate, Answer CoverageUser BehaviorTraditional SEOActive Search ClicksGEO Generative OptimizationPassive AI Answer ReceptionImpact ScopeTraditional SEOSearch Results PageGEO Generative OptimizationAI Conversations, Smart AssistantsFuture TrendsTraditional SEOGrowth StabilizingGEO Generative OptimizationExponential GrowthHear What They SayReal reviews from enterprise marketing managersZhang MingContent Marketing Director · Tech Company"After using the GEO detection tool, we found that only 30% of our articles were cited by AI platforms. Through targeted optimization, the citation rate increased to 78% within 3 months, and brand exposure in AI search tripled.“Li FangSEO Manager · E-commerce Platform"When formulating GEO strategies for clients, this tool is an essential verification method. The professional scoring report allows clients to clearly see the input-output ratio, greatly enhancing our service value.“Wang QiangMarketing Consulting Partner · Consulting Company"Previously, we only focused on Baidu rankings and completely ignored AI search as a channel. GEO detection let us see new traffic entry points, and now AI platforms bring us over 100,000 brand impressions per month.“You Might AskHow does GEO detection work?Based on the URL content you submit, we intelligently generate relevant questions, then query mainstream AI platforms (Doubao, Baidu AI, Tongyi Qianwen, DeepSeek, Tencent Yuanbao, etc.), analyze whether your links are cited in the answers, and provide a comprehensive score.How long does detection take?Single link detection is usually completed within 30 seconds to 2 minutes, depending on the response speed of each AI platform. Batch detection supports background processing, you can check progress at any time, and the system will automatically save results after completion.How to improve citation rates on AI platforms?Our reports provide detailed optimization suggestions, inclu

EN在线咨询提交需求AI Search Visibility CheckerIntelligently detect your brand’s visibility on AI platforms - Professional GEO optimization analysis toolStep 1: Select Detection PlatformsDoubaoWenxin YiyanDeepseekTongyi QianwenTencent YuanbaoStep 2: The question you ask AIRecommended Keywords0/30Step 3: Brand Keywords+Multiple hit terms please press Enter key, support up to 0/10 terms at onceStart GEO DetectionDetection PrincipleDetection PrincipleSimulate real user search scenariosAnalyze brand mentions in AI responsesCount recommendation frequency and rankingExpose competitor performance and impactDetection AdvantagesDetection AdvantagesCover 5 major AI platformsReal-time detection, accurate resultsMulti-dimensional data analysisProfessional optimization suggestionsApplication ScenariosApplication ScenariosBrand awareness monitoringCompetitor analysis comparisonGEO optimization effectiveness assessmentMarketing strategy developmentContact UsLet’s Start the New Era of AI Marketing TogetherContact us now to get exclusive GEO optimization solutions. Help your brand seize opportunities in the era of AI intelligent Q&A and achieve leapfrog business growth.Scan QR code to add human customer serviceBook a Free ConsultationNameCompany NamePhone NumberYour RequirementsSubmit BookingAbout UsXiamen Xianglian Technology Co., Ltd. was founded in 2005 and is a national high-tech enterprise with nearly 20 years of experience in the Internet service field.The company operates core platforms including Webmaster Home, Webmaster Tools, and AIBase: Webmaster Tools, as one of China’s TOP3 SEO/website operation and maintenance tool platforms, provides over 200 free tools, has served more than 5 million users, and processes tens of millions of daily queries;AIBase is positioned as China’s leading AI navigation platform, aggregating 10,000+ AI models and tools to provide one-stop AI services for developers and enterprises.Xianglian Technology is committed to providing professional, efficient technical tools and service support for Chinese Internet practitioners, helping digital transformation and development.Address：11th Floor, Building B, Feiyu Building, Huli District, Xiamen, Fujian ProvincePhone Number：18950101783Empowering the Future, Your AI Solution Knowledge BaseEnglish简体中文繁體中文にほんごCopyright © 2024-2026 厦门享联科技有限公司备案号：闽ICP备08105208号-24

EN在线咨询提交需求AI Search Engine OptimizationBrand Monitoring ServiceMonitor your brand across Google Gemini, ChatGPT, Claude & all major AI models.1Add Brand Information2Analyze Brand Keywords & Competitors3Analysis ReportMonitor by Brand DescriptionMonitor by Website AddressYour Brand NameYour Brand Description0 / 500Your Products/ServicesPlease press Enter to confirm products/services, at least 2 requiredSelect Monitoring AI PlatformSelect Monitoring Time1days7days15daysCustomizableStart AnalysisBrand Score RankingRankBrand NameScoreMeet Enterprise Data Customization NeedsCustomized ReportsAutomatically track your brand through exclusive brand reports. Track key KPIs such as brand mentions, brand coverage, or domain referencesOn-Demand SupportRely on our experienced experts to support you in building artificial intelligence search strategiesEnd-to-End SolutionsGuide optimization of enterprise/product/service pages, community sentiment, and tweet contentContact UsMost Extensive CoverageAnalyze responses from mainstream large models and AI searches at home and abroad, including: Yuanqi, MetaCat, ChatGPTComprehensive Competitive AnalysisIdentify strengths and weaknesses, find gaps to compensate for brand blind spotsGrasp Current Market HotspotsObtain real-time intelligence on all brand-related prompts, keywords, and content sourcesBrand Exclusive TrackingMonitor key metrics at all levels, from the overall brand to individual productsEmpowering the Future, Your AI Solution Knowledge BaseEnglish简体中文繁體中文にほんご© 2026AIBase

AI改变搜索引擎模式超过60%的用户开始使用AI助手查询产品和服务推荐，而不是传统搜索引擎。当用户问“最好的项目管理工具是什么？”时，你的品牌是否出现在回答中？用户提问哪个品牌的智能手表更适合运动？如果AI没有推荐你的品牌，你就失去了这个潜在客户

AI NEWSLatest AI NewsArticleSwiftlet fits an 80B-level Qwen giant into Mac: peak memory is only 4.3GB, iPhone 17 can also run 35B nativelyPublished in Latest AI NewsTime :Aug 4, 2026Read :8minuteA Swift + Metal runtime called Swiftlet is redefining “where large models can run.” It is specifically tailored for the Qwen3-Next and Qwen3.5/3.6 family of MoE (Mixture of Experts) models. The core idea is quite clever: only the small dense core of the model stays in memory, while the large part—the routing expert weights—normally resides on SSD and is streamed in as needed.The results are clear from the numbers. On an M5 Mac, the 4-bit version of Qwen3.6-35B-A3B takes up 18GB on disk but only peaks at 2.6GB of memory, decoding at 7 to 11 tok/s; even more impressive is the 4-bit version of Qwen3-Next-80B-A3B, which requires 42GB on disk but peaks at just 4.3GB of memory, with a speed of 4.5 to 5 tok/s. The 35B version can now run on an iPhone 17, with about 2.5GB of memory and a speed of around 1 tok/s — according to the developers, this is the first time such a model runs natively on a phone without touching a server.The project is fully end-to-end usable, and both models can produce verified correct outputs. The developer also admits a trade-off: each token actually activates about 3B parameters, so these models behave like large models during conversations and writing, but still act like small models in terms of factual memory.Its working principle lies in the fine-grained scheduling. Each layer routes each token to 10 out of 512 experts (for the 80B version) or 8 out of 256 experts (for the 35B version). Swiftlet keeps the dense weights — attention, DeltaNet projections, routers, shared experts, and embedding vectors — firmly in memory, taking about 1.3GB (for 35B) or 2.5GB (for 80B) in 4-bit. Thousands of routing experts are repackaged into fixed-step data blocks, stored in .qpack containers. To fetch one expert, it only needs one pread operation on the SSD, no mmap, and it doesn’t disturb the page cache. Popular experts are cached in a limited pool using LFU plus recentness strategy for eviction; the hit rate ranges from 43% to 70%, and the cache size has almost no impact on speed, as Apple’s SSD can handle the overhead of misses.The entire forward pass runs on Metal using runtime-compiled shaders, so there’s no need for a Metal toolchain during build, and the same code can be directly deployed to iOS. More interestingly, 75% of the layers use Gated DeltaNet linear attention with a fixed-size cyclic state, meaning that regardless of the context length, it never grows an expanding KV cache — the hidden burden of long text conversations is quietly offloaded.Swiftlet isn’t just a command-line toy. It has four identities designed for itself. As a library, SwiftletCore can be embedded into any macOS or iOS app, providing chat capabilities with streaming incremental output and conversation caching; as a command-line tool, swiftlet chat and swiftlet generate handle local execution and benchmarking, while swiftlet-repack builds containers directly from MLX checkpoints and supports resuming downloads from Hugging Face. As a server, swiftlet-server provides OpenAI-compatible chat-completions interfaces on a loopback address, allowing any OpenAI-compatible chat interface to connect to the local model; as an application, the iOS version of Priv AI embeds SwiftletCore as a streaming model engine, allowing regular users to start chatting by simply downloading it.Regarding correctness, every layer’s forward pass is compared layer-by-layer against the mlx-lm reference implementation, covering f32 and int4 quantization forms. Incremental decoding is also compared with full sequence results, and the Metal kernels have been repeatedly verified against precise CPU references. The container can also perform byte-level verification against the source checkpoint — whether experts are read from cache or disk, the answers are exactly the same.Its inspiration path is clearly explained: TurboFieldfare first validated the feasibility of streaming experts for Gemma on Mac, and Swiftlet borrowed some publicly available design experiences from it, such as using pread to stream experts into bounded slot pools, using LFU plus recentness for eviction, and using fixed-step packaging so that one read equals one fetch. However, the rest was written from scratch, with about 10,000 lines of Swift and Metal code, tackling the completely different Qwen mixed architecture: Gated DeltaNet linear attention, gated GQA, and high-sparse MoE with shared experts. It even implements MLX-style int4/int8 group quantization calculations in Metal, using byte-addressable kernels and 64-bit offsets to support gigabytes of shards.

AI NEWSLatest AI NewsArticleNo Cloud API Required! Liquid AI Launches Local-Running LFM2.5 Edge Intelligence ModelPublished in Latest AI NewsTime :Aug 5, 2026Read :2minuteOn August 5, the AI startup Liquid AI officially announced the release of LFM2.5-2.6B, a terminal-side model designed for agent workflows. The model has 2.6 billion parameters and can run entirely locally on terminal devices such as smartphones without relying on cloud API, providing developers with a low-cost, low-latency, and privacy-protected edge-side agent solution.In terms of technical architecture and training approach, LFM2.5-2.6B is pre-trained on about 34 trillion token, with a vocabulary expanded to 128K. Its post-training process integrates four stages: supervised fine-tuning, teacher specialization, multi-domain strategy distillation, and agent reinforcement learning, giving the model excellent planning, tool calling, and complex multi-step task handling capabilities. Benchmark tests show that the model’s performance in following instructions and using tools is comparable to large models nearly four times its size.Additionally, the model performs well in hardware compatibility and inference performance, achieving a reasoning speed of 220 tokens/second on the M5Max chip, and maintaining a stable response speed of 30 tokens/second on mobile devices. Currently, LFM2.5-2.6B and its base version are fully open-sourced on Hugging Face, and natively support mainstream inference frameworks such as llama.cpp, MLX, and vLLM, marking the practical stage of high-efficiency edge-side agents.

AI NEWSLatest AI NewsArticleAli Meoo Suiwu Launches Team Edition and Integrates Qwen-3.8-Max ModelPublished in Latest AI NewsTime :Aug 10, 2026Read :3minuteAliAI application creation platform Meoo (Miao Wu) recently launched its team version and integrated the Qwen-3.8-Max model, offering subscription services to teams and enterprise users. The new version focuses on enterprise collaboration needs, adding features such as unified identity management, a shared credit pool, a permission system, and a team skill market.Meoo was launched in April this year. Users do not need programming experience; they can generate applications, websites, H5 pages, or WeChat mini programs by simply describing their needs in natural language, and it supports one-click deployment. This team version upgrade mainly addresses enterprises’ needs in resource management, application sharing, and access control.In terms of management capabilities, enterprise members can log in using their Alibaba Cloud accounts. Teams can purchase seats and credits collectively, and administrators can allocate quotas based on member needs. Meanwhile, the platform supports centralized management of applications, skills, and cloud resources, and provides a three-level permission system: owner, administrator, and regular member.Additionally, Meoo introduced a team skill market, allowing users to package business capabilities such as pricing rules and approval processes into reusable “skills,” helping enterprises accumulate digital assets. The platform also supports custom domain deployment and added a “personalized memory” feature that can record user preferences, project rules, and business terminology.

AI NEWSLatest AI NewsArticleAI Promises a Four-Day Workweek, but Employees Work 90 Hours of Overtime per Week: Tech Giants Themselves Haven’t Achieved ItPublished in Latest AI NewsTime :Aug 10, 2026Read :5minuteTechnology giants such as Google, OpenAI, Anthropic, and Meta invest billions of dollars annually in developing AI. For years, executives have painted the same future: when AI takes over more tasks, humans can work fewer days. OpenAI has even publicly called on companies to trial a four-day workweek without pay cuts, claiming AI will greatly increase labor efficiency. However, a BBC investigation found that these companies promoting shorter working hours have never lived up to their promises. Former OpenAI employees revealed that during their tenure, they worked at least 70 hours per week, and during peak periods, even over 90 hours, with weekends still being used to catch up or troubleshoot system failures.Ironically, the more AI increases productivity, the longer employees’ working hours become. Meta recently laid off about 10% of its staff while forcing many into AI project teams, which employees privately call “conscription”—refusing means resigning. After joining an AI team, working late into the night and continuing on weekends has become the norm, and even after work, they remain on standby. A former Google employee said that after the company continuously shifted processing power and storage resources to AI projects, internal engineering system failures became more frequent, often requiring them to stay up late to resolve issues. After leaving, their sleep and health conditions significantly improved.The time saved by AI is consumed by new workA study by the University of California, Berkeley, tracking hundreds of employees at a tech company for eight months revealed a core paradox: AI makes work faster and tasks broader, but the amount of time spent working each day actually gets longer. A key reason is that employees must constantly check whether AI-generated content is correct, a “supervision cost” that offsets the efficiency gains from AI. MIT innovation researchers pointed out that even if AI does save time, the saved portion will be swallowed by the work of system adjustments, implementation, and ensuring smooth operations.A deeper driving force comes from employees’ own anxiety. Some want to use the time saved to complete more work, while others fear that not continuously increasing output may prevent them from proving their value to employers. New work will always emerge, keeping the idea that “a 20% reduction in workload could lead to a four-day workweek” stuck in theory. Anthrop’s CEO, Amodei, has openly stated that the higher the AI productivity, the more jobs may be eliminated in the future, further increasing employees’ pressure to not relax. AI was supposed to do many things for humans, but in reality, at least in the core areas of the tech industry, more people are not sleeping better or living more easily, but instead falling into a cycle where the more efficient they are, the busier they become.