[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-bluffjax-open-source-suite-trains-poker-playing-ai-on-gpus":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},10588,"bluffjax-open-source-suite-trains-poker-playing-ai-on-gpus","BluffJAX open-source suite trains poker-playing AI on GPUs","A new open-source JAX library lets researchers train poker-playing AI at hundreds of millions of samples per second on GPUs.","A team of researchers just open-sourced a toolkit for training AI to bluff, fold, and call at GPU speed.\n\nBluffJAX is a suite of adversarial, imperfect-information card games built in JAX, designed to run reinforcement learning simulations on GPU accelerators. It includes well-known benchmarks like Texas Hold'Em and Kuhn Poker, plus games such as Bluff, Stud Poker, and Kemps that haven't previously been used in RL research. The researchers say the suite scales to hundreds of millions of samples per second across single and multi-GPU setups. They also benchmarked reinforcement learning, tree search, and game-solving algorithms against the suite, giving other researchers a baseline for comparison.\n\nPoker has long been a proving ground for AI that has to reason under uncertainty - bluffing, hidden information, no clear read on what an opponent holds. Most existing tools for this kind of research run on CPUs or smaller GPU libraries, which caps how fast and how large experiments can get. Building on JAX's GPU-native parallelism means more training iterations in the same stretch of time, and sample count is usually the thing reinforcement learning research is starved for.\n\nWhether any of this produces better real-world decision-making AI - the pitch behind poker-bot research since Libratus and Pluribus - is a separate question; for now, it's mostly a faster box of toys for academics.","[\"ai\",\"reinforcement-learning\",\"open-source\",\"game-theory\"]","2026-10-07T04:00:00.000Z","2026-10-08T21:24:11.707Z","2026-10-08T21:24:14.104Z","published",null,[],"ai",[24,26,27,28],"reinforcement-learning","open-source","game-theory",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.07686",0,{"sections":35},[36,40,45,50,55,60,65,70,75,79,84,89,94,99],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",6448,"2026-10-07T18:45:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",904,"2026-10-07T19:53:42.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Policy","policy",474,"2026-10-07T18:23:21.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Deals","deals",453,"2026-10-07T23:58:31.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",222,"2026-10-07T21:19:54.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Science","science",186,"2026-10-06T21:20:39.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Consumer Tech","consumer-tech",174,"2026-10-07T17:41:41.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Software","software",113,"2026-10-07T18:10:00.000Z",{"name":76,"slug":77,"count":73,"latest_published_at":78},"Startups","startups","2026-10-07T23:36:57.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Dev Tools","dev-tools",105,"2026-10-07T16:59:11.000Z",{"name":85,"slug":86,"count":87,"latest_published_at":88},"General","general",61,"2026-10-07T22:00:24.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"Gaming","gaming",56,"2026-10-07T12:00:00.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",33,"2026-10-05T11:57:17.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",8,"2026-10-05T09:00:00.000Z"]