[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"nav-categories":3,"article-xai-s-efficiency-crisis-11-mfu-a-founder-exodus-and-a-60-billion-hail-mary-on-cursor":70},{"data":4},[5,37,57,64],{"name":6,"slug":7,"categories":8},"Productivity","productivity",[9,13,17,21,25,29,33],{"id":10,"title":11,"slug":12},17,"Branding","branding",{"id":14,"title":15,"slug":16},19,"Marketing","marketing",{"id":18,"title":19,"slug":20},20,"Work","work",{"id":22,"title":23,"slug":24},34,"Community","community",{"id":26,"title":27,"slug":28},21,"For newbies","for-newbies",{"id":30,"title":31,"slug":32},24,"Investment","investment",{"id":34,"title":35,"slug":36},22,"Finance","finance",{"name":38,"slug":39,"categories":40},"Tech","tech",[41,45,49,53],{"id":42,"title":43,"slug":44},28,"Technology","technology",{"id":46,"title":47,"slug":48},32,"Artificial Intelligence","artificial-intelligence",{"id":50,"title":51,"slug":52},26,"Security and protection","security-and-protection",{"id":54,"title":55,"slug":56},31,"YouTube Blog","youtube-blog",{"name":58,"slug":59,"categories":60},"News","news",[61],{"id":62,"title":58,"slug":63},18,"quasanews",{"name":65,"slug":66,"categories":67},"Business","business",[68],{"id":69,"title":65,"slug":66},16,{"post":71,"published_news":95,"popular_news":149,"categories":219},{"title":72,"description":73,"meta_title":72,"meta_description":74,"meta_keywords":75,"text":76,"slug":77,"created_at":78,"publish_at":78,"formatted_created_at":79,"category_id":62,"links":80,"view_type":85,"video_url":86,"views":34,"likes":87,"lang":88,"comments_count":87,"category":89},"xAI’s Efficiency Crisis: 11% MFU, a Founder Exodus, and a $60 Billion Hail Mary on Cursor","Internal numbers that just leaked show the company is running its massive Colossus supercluster at a shocking 11% Model FLOPS Utilization (MFU) during training.","Internal numbers that just leaked show the company is running its massive Colossus supercluster at a shocking 11% Model FLOPS Utilization (MFU) during training.","For context, well-run frontier labs typically hit 30–45% on pre-training runs. xAI isn’t inventing a radically new architecture — it’s just struggling to keep the GPUs busy.","\u003Cp>It’s been a rough few months for xAI.\u003C/p>\n\n\u003Cp>\u003Cpicture>\u003Csource srcset=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191739879.webp\" type=\"image/webp\">\u003Cimg alt=\"xAI’s Efficiency Crisis: 11% MFU, a Founder Exodus, and a  Billion Hail Mary on Cursor\" class=\"image-align-left\" height=\"372\" src=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191739879.jpg\" width=\"250\" />\u003C/picture>Internal numbers that just leaked show the company is running its massive Colossus supercluster at a shocking \u003Cstrong>11% Model FLOPS Utilization (MFU)\u003C/strong> during training. For context, well-run frontier labs typically hit 30–45% on pre-training runs. xAI isn’t inventing a radically new architecture — it’s just struggling to keep the GPUs busy.\u003C/p>\n\n\u003Cp>That single number explains a lot. To train a model of any given size, xAI effectively needs \u003Cstrong>three times\u003C/strong> the GPUs or three times the time compared to its better-optimized peers. All the bragging about the world’s largest cluster suddenly looks a lot less impressive when the iron is sitting idle most of the time.\u003C/p>\n\n\u003Cp>The company has now publicly set an ambitious internal target of \u003Cstrong>50% MFU\u003C/strong> — classic Elon over-the-top goal-setting. Whether they hit it or not, the current reality is clear: Colossus is dramatically underperforming.\u003C/p>\n\n\u003Chr />\n\u003Ch4>\u003Cstrong>When You Can’t Use It, Rent It\u003C/strong>\u003C/h4>\n\n\u003Cp>\u003Ca href=\"https://www.theinformation.com/briefings/xai-rent-computing-power-cursor\">The situation got so bad that SpaceX (which absorbed xAI earlier this year) started doing something pragmatic and slightly humiliating\u003C/a>: \u003Cstrong>renting out chunks of Colossus to third parties\u003C/strong>.\u003C/p>\n\n\u003Cp>The biggest tenant so far? AI coding startup \u003Cstrong>Cursor\u003C/strong>, which is using tens of thousands of xAI GPUs to train its next-generation models (Composer 2.5 and beyond). In other words: “We can’t squeeze enough performance out of our own hardware to stay competitive, so we’re selling cycles to people who can.”\u003C/p>\n\n\u003Cp>The message is blunt: if xAI can’t make the iron sing for Grok, at least it can make money while someone else does.\u003C/p>\n\n\u003Chr />\n\u003Ch4>\u003Cstrong>The Bigger Picture Isn’t Pretty\u003C/strong>\u003C/h4>\n\n\u003Cp>\u003Cpicture>\u003Csource srcset=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191741965.webp\" type=\"image/webp\">\u003Cimg alt=\"xAI’s Efficiency Crisis: 11% MFU, a Founder Exodus, and a  Billion Hail Mary on Cursor\" class=\"image-align-right\" height=\"372\" src=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191741965.jpg\" width=\"250\" />\u003C/picture>The efficiency problem is only part of the story.\u003C/p>\n\n\u003Cp>\u003Cstrong>Over the past six months:\u003C/strong>\u003C/p>\n\n\u003Cul>\n\t\u003Cli>\u003Cstrong>Every co-founder except Elon has left\u003C/strong> xAI.\u003C/li>\n\t\u003Cli>There hasn’t been a single notable frontier-level LLM release from the company in over half a year.\u003C/li>\n\t\u003Cli>Revenue growth is anemic while OpenAI, Anthropic, and others are seeing explosive demand.\u003C/li>\n\u003C/ul>\n\n\u003Cp>Even if xAI somehow ships a competitive model tomorrow, it risks hitting the same wall Anthropic did: great model, nowhere near enough inference capacity to serve customers at scale. Start stealing GPUs from training runs to meet inference demand, and the next training run falls further behind. Classic death spiral.\u003C/p>\n\n\u003Chr />\n\u003Ch4>\u003Cstrong>Elon’s Knight Move: The $60 Billion Cursor Option\u003C/strong>\u003C/h4>\n\n\u003Cp>\u003Cpicture>\u003Csource srcset=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191744262.webp\" type=\"image/webp\">\u003Cimg alt=\"xAI’s Efficiency Crisis: 11% MFU, a Founder Exodus, and a  Billion Hail Mary on Cursor\" class=\"image-align-left\" height=\"447\" src=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191744262.jpg\" width=\"300\" />\u003C/picture>Rather than wait for internal fixes, Elon has gone full chess grandmaster.\u003C/p>\n\n\u003Cp>SpaceX just announced a partnership with Cursor that includes \u003Cstrong>an option to acquire the entire company for $60 billion later this year\u003C/strong>.\u003C/p>\n\n\u003Cp>If the collaboration on “the world’s best coding and knowledge-work AI” delivers, SpaceX can pull the trigger. If it doesn’t… they walk away after paying a \u003Cstrong>$10 billion breakup fee\u003C/strong> for the joint work.\u003C/p>\n\n\u003Cp>It’s an elegant (and very expensive) way to inject serious coding talent and distribution into the xAI/SpaceX orbit while getting immediate access to real usage data and engineering momentum.\u003C/p>\n\n\u003Cp>Cursor gets the compute it desperately needs.\u003C/p>\n\n\u003Cp>SpaceX/xAI gets a potential fast path to a killer coding agent that could actually move the revenue needle.\u003C/p>\n\n\u003Cp>Also read:\u003C/p>\n\n\u003Cul>\n\t\u003Cli>\u003Ca href=\"https://quasa.io/media/llms-explained-how-large-language-models-work-and-where-they-commonly-fail\">LLMs Explained: How Large Language Models Work (and Where They Commonly Fail)\u003C/a>\u003C/li>\n\t\u003Cli>\u003Ca href=\"https://quasa.io/media/youtube-s-new-push-notification-crackdown-a-smart-fix-for-notification-fatigue-or-a-hit-to-creator-reach\">YouTube’s New Push Notification Crackdown: A Smart Fix for Notification Fatigue or a Hit to Creator Reach?\u003C/a>\u003C/li>\n\t\u003Cli>\u003Ca href=\"https://quasa.io/media/the-invisible-front-and-big-money-how-caci-is-conquering-the-world-of-electronic-warfare\">The Invisible Front and Big Money: How CACI is Conquering the World of Electronic Warfare\u003C/a>\u003C/li>\n\u003C/ul>\n\n\u003Chr />\n\u003Ch4>\u003Cstrong>Last Convulsion or Master Stroke?\u003C/strong>\u003C/h4>\n\n\u003Cp>\u003Cpicture>\u003Csource srcset=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191747379.webp\" type=\"image/webp\">\u003Cimg alt=\"xAI’s Efficiency Crisis: 11% MFU, a Founder Exodus, and a  Billion Hail Mary on Cursor\" class=\"image-align-right\" height=\"298\" src=\"https://cdn.quasa.io/photos/0001/image-2026-04-23t191747379.jpg\" width=\"200\" />\u003C/picture>This feels like xAI’s (or rather, SpaceX’s) last serious swing at staying in the frontier race before the IPO window closes.\u003C/p>\n\n\u003Cul>\n\t\u003Cli>\u003Cstrong>Best case:\u003C/strong> The Cursor partnership delivers a breakout coding model, utilization climbs, revenue finally accelerates, and the combined entity rides a successful IPO to the next level.\u003C/li>\n\t\u003Cli>\u003Cstrong>Worst case:\u003C/strong> The $10 billion (or full $60 billion) price tag becomes an expensive reminder that hardware alone isn’t enough — you also need software, people, and execution.\u003C/li>\n\u003C/ul>\n\n\u003Cp>Right now, the market is watching closely. xAI built the biggest cluster on Earth. The question everyone is asking is no longer “How many GPUs do you have?” but the far more uncomfortable one: \u003Cstrong>How well are you actually using them?\u003C/strong>\u003C/p>\n\n\u003Cp>And right now, the answer is: not well enough.\u003C/p>","xai-s-efficiency-crisis-11-mfu-a-founder-exodus-and-a-60-billion-hail-mary-on-cursor","2026-04-23T17:24:47.000000Z","23.04.2026",{"image":81,"image_webp":82,"thumb":83,"thumb_webp":84},"https://cdn.quasa.io/images/news/UdcHExtyqwxs3tp6asCqX8nGeIRetAlx5672Ht5g.jpg","https://cdn.quasa.io/images/news/UdcHExtyqwxs3tp6asCqX8nGeIRetAlx5672Ht5g.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/UdcHExtyqwxs3tp6asCqX8nGeIRetAlx5672Ht5g.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/UdcHExtyqwxs3tp6asCqX8nGeIRetAlx5672Ht5g.webp","small",null,0,"en",{"id":62,"title":58,"slug":63,"meta_title":90,"meta_description":91,"meta_keywords":92,"deleted_at":86,"created_at":93,"updated_at":94,"lang":88},"Quasa News: Latest Updates, AI & Tech Announcements | QUASA","Daily news and deep analysis on AI, cryptocurrencies, and key business trends. Find out what's next for the markets. Stay informed with QUASA.","AI news, crypto news, blockchain news, IT news, market analysis, business trends, latest technology news today, artificial intelligence news, major crypto events, future of tech news","2021-08-09T19:09:44.000000Z","2026-04-22T15:16:57.000000Z",[96,109,112,124,136],{"title":97,"description":98,"slug":99,"created_at":100,"publish_at":100,"formatted_created_at":79,"category":101,"links":102,"view_type":85,"video_url":86,"views":107,"likes":87,"lang":88,"comments_count":87,"is_pinned":108},"Introducing GPT-5.5: A New Class of Intelligence for Real Work and Powering Agents","OpenAI has officially launched GPT-5.5 — the latest and most capable model in its flagship series — and it arrives with a clear mission: to move artificial intelligence beyond clever conversation and into the realm of reliable, goal-oriented work.","introducing-gpt-5-5-a-new-class-of-intelligence-for-real-work-and-powering-agents","2026-04-23T20:21:38.000000Z",{"title":47,"slug":48},{"image":103,"image_webp":104,"thumb":105,"thumb_webp":106},"https://cdn.quasa.io/images/news/cquBkjqHU4PeYGOyUJms5Ozmg2OYlggNlEj8zlfj.jpg","https://cdn.quasa.io/images/news/cquBkjqHU4PeYGOyUJms5Ozmg2OYlggNlEj8zlfj.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/cquBkjqHU4PeYGOyUJms5Ozmg2OYlggNlEj8zlfj.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/cquBkjqHU4PeYGOyUJms5Ozmg2OYlggNlEj8zlfj.webp",5,false,{"title":72,"description":73,"slug":77,"created_at":78,"publish_at":78,"formatted_created_at":79,"category":110,"links":111,"view_type":85,"video_url":86,"views":34,"likes":87,"lang":88,"comments_count":87,"is_pinned":108},{"title":58,"slug":63},{"image":81,"image_webp":82,"thumb":83,"thumb_webp":84},{"title":113,"description":114,"slug":115,"created_at":116,"publish_at":116,"formatted_created_at":79,"category":117,"links":118,"view_type":123,"video_url":86,"views":50,"likes":87,"lang":88,"comments_count":87,"is_pinned":108},"LLMs Explained: How Large Language Models Work (and Where They Commonly Fail)","An LLM is a statistical system trained to predict the next token in a sequence. The surprising part is how far “next token prediction” can go when you scale data, model size, and training compute.","llms-explained-how-large-language-models-work-and-where-they-commonly-fail","2026-04-23T16:37:41.000000Z",{"title":47,"slug":48},{"image":119,"image_webp":120,"thumb":121,"thumb_webp":122},"https://cdn.quasa.io/images/news/1nHfaaxNxTekFwzUBoKsQE8ZEQIgD44Xvwu8ZrSB.jpg","https://cdn.quasa.io/images/news/1nHfaaxNxTekFwzUBoKsQE8ZEQIgD44Xvwu8ZrSB.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/1nHfaaxNxTekFwzUBoKsQE8ZEQIgD44Xvwu8ZrSB.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/1nHfaaxNxTekFwzUBoKsQE8ZEQIgD44Xvwu8ZrSB.webp","large",{"title":125,"description":126,"slug":127,"created_at":128,"publish_at":128,"formatted_created_at":79,"category":129,"links":130,"view_type":85,"video_url":86,"views":135,"likes":87,"lang":88,"comments_count":87,"is_pinned":108},"YouTube’s New Push Notification Crackdown: A Smart Fix for Notification Fatigue or a Hit to Creator Reach?","In the ongoing conversation around content deliverability, YouTube has just rolled out a significant change that will affect millions of creators and subscribers.","youtube-s-new-push-notification-crackdown-a-smart-fix-for-notification-fatigue-or-a-hit-to-creator-reach","2026-04-23T12:32:15.000000Z",{"title":55,"slug":56},{"image":131,"image_webp":132,"thumb":133,"thumb_webp":134},"https://cdn.quasa.io/images/news/CeKKOmW8G5UoYDFB91yqL5J8iyEnOm9Uqw3OPNv1.jpg","https://cdn.quasa.io/images/news/CeKKOmW8G5UoYDFB91yqL5J8iyEnOm9Uqw3OPNv1.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/CeKKOmW8G5UoYDFB91yqL5J8iyEnOm9Uqw3OPNv1.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/CeKKOmW8G5UoYDFB91yqL5J8iyEnOm9Uqw3OPNv1.webp",48,{"title":137,"description":138,"slug":139,"created_at":140,"publish_at":141,"formatted_created_at":79,"category":142,"links":143,"view_type":85,"video_url":86,"views":148,"likes":87,"lang":88,"comments_count":87,"is_pinned":108},"The Rice on the Chessboard: Why Humanity Keeps Underestimating Exponential AI Growth","There’s an ancient Indian parable about a wise man who, for a service rendered, asked a powerful king to pay him in rice — placing one grain on the first square of a chessboard, two on the second, four on the third, and doubling the amount with every subsequent square.","the-rice-on-the-chessboard-why-humanity-keeps-underestimating-exponential-ai-growth","2026-04-20T20:10:17.000000Z","2026-04-23T11:00:00.000000Z",{"title":47,"slug":48},{"image":144,"image_webp":145,"thumb":146,"thumb_webp":147},"https://cdn.quasa.io/images/news/A7vLK20XbrRVdUCdvtedODguBAKoaZY79tMBlsOE.jpg","https://cdn.quasa.io/images/news/A7vLK20XbrRVdUCdvtedODguBAKoaZY79tMBlsOE.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/A7vLK20XbrRVdUCdvtedODguBAKoaZY79tMBlsOE.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/A7vLK20XbrRVdUCdvtedODguBAKoaZY79tMBlsOE.webp",58,[150,163,178,190,205],{"title":151,"description":152,"slug":153,"created_at":154,"publish_at":155,"formatted_created_at":156,"category":157,"links":158,"view_type":85,"video_url":86,"views":161,"likes":162,"lang":88,"comments_count":87,"is_pinned":108},"The Anatomy of an Entrepreneur","Entrepreneur is a French word that means an enterpriser. Enterprisers are people who undertake a business or enterprise with the chance of earning profits or suffering from loss.","the-anatomy-of-an-entrepreneur","2021-08-04T15:18:21.000000Z","2025-12-14T06:09:00.000000Z","14.12.2025",{"title":65,"slug":66},{"image":159,"image_webp":86,"thumb":160,"thumb_webp":160},"https://cdn.quasa.io/images/news/mVsXPTMuHZuI7UXCsENgL1Qwp1uSOf7Rz3uVPMfm.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/mVsXPTMuHZuI7UXCsENgL1Qwp1uSOf7Rz3uVPMfm.webp",71355,2,{"title":164,"description":165,"slug":166,"created_at":167,"publish_at":168,"formatted_created_at":169,"category":170,"links":171,"view_type":123,"video_url":86,"views":176,"likes":177,"lang":88,"comments_count":87,"is_pinned":108},"Advertising on QUASA","QUASA MEDIA is read by more than 400 thousand people a month. We offer to place your article, add a link or order the writing of an article for publication.","advertising-on-quasa","2022-07-06T07:33:02.000000Z","2025-12-15T17:33:02.000000Z","15.12.2025",{"title":58,"slug":63},{"image":172,"image_webp":173,"thumb":174,"thumb_webp":175},"https://cdn.quasa.io/images/news/45SvmdsTQbiyc3nxgbyHY1mpVbisYyub2BCHjqBL.jpg","https://cdn.quasa.io/images/news/45SvmdsTQbiyc3nxgbyHY1mpVbisYyub2BCHjqBL.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/45SvmdsTQbiyc3nxgbyHY1mpVbisYyub2BCHjqBL.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/45SvmdsTQbiyc3nxgbyHY1mpVbisYyub2BCHjqBL.webp",71128,4,{"title":179,"description":180,"slug":181,"created_at":182,"publish_at":183,"formatted_created_at":184,"category":185,"links":186,"view_type":85,"video_url":86,"views":189,"likes":177,"lang":88,"comments_count":87,"is_pinned":108},"What is a Startup?","A startup is not a new company, not a tech company, nor a new tech company. You can be a new tech company, if your goal is not to grow high and fast; then, you are not a startup. ","what-is-a-startup","2021-08-04T12:05:17.000000Z","2025-12-17T13:02:00.000000Z","17.12.2025",{"title":65,"slug":66},{"image":187,"image_webp":86,"thumb":188,"thumb_webp":188},"https://cdn.quasa.io/images/news/EOsQhSW3VXyG7a6NPdE1oZd00xfJXe3bjY5aJGb7.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/EOsQhSW3VXyG7a6NPdE1oZd00xfJXe3bjY5aJGb7.webp",68745,{"title":191,"description":192,"slug":193,"created_at":194,"publish_at":195,"formatted_created_at":196,"category":197,"links":198,"view_type":85,"video_url":86,"views":203,"likes":162,"lang":88,"comments_count":204,"is_pinned":108},"Top 5 Tips to Make More Money as a Content Creator","Content creators are one of the most desired job titles right now. Who wouldn’t want to earn a living online?","top-5-tips-to-make-more-money-as-a-content-creator","2022-01-17T17:31:51.000000Z","2026-01-17T11:30:00.000000Z","17.01.2026",{"title":19,"slug":20},{"image":199,"image_webp":200,"thumb":201,"thumb_webp":202},"https://cdn.quasa.io/images/news/gP8kiumBPpJmQv6SMieXiX1tDetx43VwFfO1P4Ca.jpg","https://cdn.quasa.io/images/news/gP8kiumBPpJmQv6SMieXiX1tDetx43VwFfO1P4Ca.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/gP8kiumBPpJmQv6SMieXiX1tDetx43VwFfO1P4Ca.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/gP8kiumBPpJmQv6SMieXiX1tDetx43VwFfO1P4Ca.webp",42690,1,{"title":206,"description":207,"slug":208,"created_at":209,"publish_at":210,"formatted_created_at":211,"category":212,"links":213,"view_type":123,"video_url":86,"views":218,"likes":162,"lang":88,"comments_count":87,"is_pinned":108},"8 Logo Design Tips for Small Businesses","Your logo tells the story of your business and the values you stand for.","8-logo-design-tips-for-small-businesses","2021-12-04T21:59:52.000000Z","2025-05-05T03:30:00.000000Z","05.05.2025",{"title":15,"slug":16},{"image":214,"image_webp":215,"thumb":216,"thumb_webp":217},"https://cdn.quasa.io/images/news/Wbx2NtS1CnTupgoQbpFMGspJ5jm4uob2hDOq33r0.jpg","https://cdn.quasa.io/images/news/Wbx2NtS1CnTupgoQbpFMGspJ5jm4uob2hDOq33r0.webp","https://cdn.quasa.io/thumbs/news-thumb/images/news/Wbx2NtS1CnTupgoQbpFMGspJ5jm4uob2hDOq33r0.jpg","https://cdn.quasa.io/thumbs/news-thumb/images/news/Wbx2NtS1CnTupgoQbpFMGspJ5jm4uob2hDOq33r0.webp",41758,[220,221,222,223,224,225,226,227,228,229,230,231,232],{"title":23,"slug":24},{"title":47,"slug":48},{"title":55,"slug":56},{"title":43,"slug":44},{"title":51,"slug":52},{"title":31,"slug":32},{"title":35,"slug":36},{"title":27,"slug":28},{"title":19,"slug":20},{"title":15,"slug":16},{"title":58,"slug":63},{"title":11,"slug":12},{"title":65,"slug":66}]