{"id":62362,"date":"2026-11-15T21:03:00","date_gmt":"2026-11-15T18:03:00","guid":{"rendered":"https:\/\/infolight.ua\/2026\/11\/15\/gemini-4-krypton-what-is-actually-known\/"},"modified":"2026-11-15T21:03:00","modified_gmt":"2026-11-15T18:03:00","slug":"gemini-4-krypton-what-is-actually-known","status":"publish","type":"post","link":"https:\/\/infolight.ua\/en\/2026\/11\/15\/gemini-4-krypton-what-is-actually-known\/","title":{"rendered":"Gemini 4 Krypton: what is actually known about Google\u2019s next \u201cheavy\u201d AI model \u2014 and what the viral leak gets wrong"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\"><strong>A widely shared post claims Google is about to release Gemini 4 Krypton, the top of a four-tier lineup \u201cbuilt along the periodic table\u201d: Hydrogen, Carbon, Argon and Krypton. We checked the claims against Google\u2019s official announcements, media reports and posts by early testers. Krypton does exist as a test checkpoint, but most of the lineup story does not match the facts.<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The text has been circulating in AI-focused Telegram channels, including <a href=\"https:\/\/t.me\/cgevent\">cgevent<\/a>. It says Gemini 4 models are ranked by the \u201cweight\u201d of the element: Hydrogen is the fastest and cheapest, Carbon a balanced all-rounder, Argon a deep-reasoning model, and Krypton a heavyweight flagship built to take on Anthropic\u2019s Claude Fable 5.2 and OpenAI\u2019s GPT-6.2 Sol. All four are supposedly tied together by a dynamic router that sends each request to the right model.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Confirmed: Argon and Carbon are already out<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Gemini 4 Argon<\/strong> is not a mid-tier model. It is the first and so far the only generally available Gemini 4 model. Google <a href=\"https:\/\/blog.google\/innovation-and-ai\/models-and-research\/gemini-models\/gemini-4-argon\/\">announced it on 30 September<\/a>, initially for cyber defenders in its Fairwind programme. Since 13 October developers can use it through the Gemini API at <strong>$2 per million input tokens and $10 per million output tokens<\/strong>, <a href=\"https:\/\/thenewstack.io\/gemini-4-argon-live\/\"><strong>The New Stack<\/strong><\/a> reports. Google calls it a flagship model for software engineering, maths, science and agent workflows.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Gemini 4 Carbon<\/strong> is not a \u201cbalanced all-rounder below Argon\u201d either. Google <a href=\"https:\/\/blog.google\/innovation-and-ai\/models-and-research\/gemini-models\/introducing-gemini-4-carbon\/\">introduced it on 28 October<\/a> as its strongest model for cybersecurity and long-running tasks, built on Argon. Access is still limited to selected cyber defenders. The independent Artificial Analysis index ranks Carbon first with <strong>59 points<\/strong>, ahead of Argon.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What we know about Krypton and Hydrogen<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Google has not announced <strong>Krypton<\/strong>. On 9 November a tester known as chetaslua posted on X a side-by-side run of Krypton, Carbon and Argon on the same prompt and called it \u201cthe first checkpoint of krypton-a\u201d. He said Krypton took about twice as long and wrote about twice as much code. Another model tracker, <a href=\"https:\/\/x.com\/koltregaskes\/status\/2119678120924201150\">Kol Tregaskes<\/a>, wrote that Krypton \u201cis real\u201d and sounds like a big model. It later surfaced in blind-test arenas.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Opinions differ: some testers call Krypton the strongest of the three, others say the gain over Carbon is small. No benchmark numbers have been published. Release dates in circulation (25 November and 8 December) come from Reddit posts, and prices are users\u2019 guesses, according to an overview by <a href=\"https:\/\/kie.ai\/blog\/what-is-gemini-4-krypton\"><strong>Kie.ai<\/strong><\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Hydrogen<\/strong> is the reported codename of a fast, low-cost Flash-tier model. X users first mentioned it in late October, citing several test checkpoints. Google has not confirmed it, and it is not listed in the Gemini API documentation.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What the viral lineup gets wrong<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u201cThe heavier the element, the more powerful the model\u201d does not hold.<\/strong> Carbon (atomic number 6) is lighter than argon (18), yet Carbon came later and, by Google\u2019s own figures, outperforms Argon.<\/li>\n\n\n\n<li><strong>The family is not a closed set of four.<\/strong> Internal documents seen by <a href=\"https:\/\/www.businessinsider.com\/google-employees-test-new-gemini-4-model-argon-barium-carbon-2026-10\">Business Insider<\/a> also list a model called Barium; its Barium-B version is what shipped as Argon. Google\u2019s internal codenames usually differ from public names.<\/li>\n\n\n\n<li><strong>The model roles and the \u201cinference-compute scaling principle\u201d have no source.<\/strong> Google has announced nothing of the kind, and the roles given to Argon and Carbon contradict the official ones.<\/li>\n\n\n\n<li><strong>A dynamic router across all four models is not confirmed anywhere.<\/strong><\/li>\n\n\n\n<li><strong>One of the \u201cdirect rivals\u201d does not exist yet.<\/strong> OpenAI has not released GPT-6.2 Sol; it is known only from rumours, and OpenAI\u2019s documentation still lists GPT-6.1 Sol as the latest Sol model.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">How to read leaks like this<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Stories about new AI models often grow out of a few posts on X: an arena screenshot, one tester\u2019s impressions, a codename. Someone then assembles them into a tidy \u201ccompany plan\u201d with roles, prices and dates \u2014 and the tidiness is what makes it convincing. Check for an official announcement, a listing in the API documentation and independent tests. Until those appear, Gemini 4 Krypton is a promising test checkpoint, not a flagship with known specifications.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>Photo: Google headquarters in Mountain View, 2009. <a href=\"https:\/\/commons.wikimedia.org\/wiki\/File:Google_1600_Amphitheatre_(4136750386).jpg\">H\u00e5kan Dahlstr\u00f6m \/ Wikimedia Commons<\/a>, <a href=\"https:\/\/creativecommons.org\/licenses\/by\/2.0\/\">CC BY 2.0<\/a><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>A viral post describes a neat four-model Gemini 4 lineup named after chemical elements. We checked it: Krypton is real as a test checkpoint, but much of the lineup contradicts what Google has already announced.<\/p>\n","protected":false},"author":25,"featured_media":62359,"comment_status":"open","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"wds_primary_category":1307,"footnotes":""},"categories":[1266,1690,1307],"tags":[],"class_list":["post-62362","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-infohihiiena","category-neus","category-technology"],"jetpack_featured_media_url":"https:\/\/infolight.ua\/wp-content\/uploads\/2026\/11\/google-kampus-mountin-viu.jpg","_links":{"self":[{"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/posts\/62362","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/users\/25"}],"replies":[{"embeddable":true,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/comments?post=62362"}],"version-history":[{"count":0,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/posts\/62362\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/media\/62359"}],"wp:attachment":[{"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/media?parent=62362"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/categories?post=62362"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infolight.ua\/en\/wp-json\/wp\/v2\/tags?post=62362"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}