--- title: xAI just caught up (Grok 4.6 is here) | daily.dev description: Grok 4.6 from xAI is reviewed as a post-training upgrade over Grok 4.5, focused on long-running agentic tasks, coding, and visual/design work. It scores 61 on... canonical: https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: xAI just caught up (Grok 4.6 is here) | daily.dev og:description: Grok 4.6 from xAI is reviewed as a post-training upgrade over Grok 4.5, focused on long-running agentic tasks, coding, and visual/design work. It scores 61 on... og:url: https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych og:image: https://api.daily.dev/og/posts/PGnjC1Ych.png og:image:alt: xAI just caught up (Grok 4.6 is here) og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # xAI just caught up (Grok 4.6 is here) **[Theo - t3․gg](https://daily.dev/sources/t3dotgg)** · 25 min read · 2 upvotes · 0 comments ## Summary Grok 4.6 from xAI is reviewed as a post-training upgrade over Grok 4.5, focused on long-running agentic tasks, coding, and visual/design work. It scores 61 on the Artificial Analysis Intelligence Index, roughly tying GPT-5.1 (referred to obliquely) and trailing Claude Opus 4.5 and Gemini 3. Benchmarks like Cursor Bench and Frontier Code show large jumps versus 4.5. However, the model got notably more expensive and slower than 4.5 because token efficiency dropped over 30%, pushing it off the ideal cost-intelligence sweet spot. Real-world testing found weak UI/design output and a failed 3D game port, though a security audit and a PR-generation task for a coding tool went reasonably well. The reviewer is unimpressed enough to keep using other models day-to-day but is optimistic about the upcoming Grok 4.7. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is different about Grok 4.6 compared to Grok 4.5? Grok 4.6 is a new post-training pass on the same Grok 4.5 pre-trained base, not a new pretraining run, focused on long-running agentic tasks and visual/interactive work. It gained roughly five points on the Artificial Analysis Intelligence Index over 4.5, jumping benchmarks like Frontier Code from 56.6 to 61.3, but token usage per run rose over 30%, making it slower and more expensive per task than 4.5. _Track how fast-moving model updates like this shift your agentic coding costs on daily.dev._ ### How much does Grok 4.6 cost compared to Claude Opus 4.5 and GPT-5.1? Grok 4.6 costs $2 per million input tokens and $6 per million output tokens, about 60% below Claude Opus 4.5's pricing and still cheaper than GPT-5.1. Real-world cost per task is around 84 cents, similar to Kimi K2, though this is more than double what Grok 4.5 cost due to lower token efficiency in the new version. _Compare frontier model pricing shifts like this before picking a coding agent on daily.dev._ ### Is Grok 4.6 good at 3D game development or design tasks? No, Grok 4.6 performed poorly on 3D and design tasks in hands-on testing. It failed outright to render a 3D scene on a first attempt (producing a black screen), and after a fix attempt it shipped inverted controls and badly misplaced ground objects, ranking as the weakest 3D output tested among current models; design mockups were also judged generic and less polished than Claude or GPT-5.1 outputs. _Weigh model-specific weaknesses like this when choosing an agent for UI or game work via daily.dev._ --- Tags: [#ai](https://daily.dev/tags/ai), [#ai-agents](https://daily.dev/tags/ai-agents), [#grok](https://daily.dev/tags/grok) [View this post on daily.dev](https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"xAI just caught up (Grok 4.6 is here)","url":"https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych"},"datePublished":"2026-08-13T11:41:45.674Z","dateModified":"2026-08-13T16:23:24.545Z","description":"Grok 4.6 from xAI is reviewed as a post-training upgrade over Grok 4.5, focused on long-running agentic tasks, coding, and visual/design work. It scores 61 on...","image":"https://i.ytimg.com/vi/c7W8jpsjtCc/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/c7W8jpsjtCc/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Theo - t3․gg","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Theo - t3․gg","logo":"https://media.daily.dev/image/upload/s--UmX7IyU3--/f_auto/v1704628081/logos/t3dotgg.jpg","url":"https://daily.dev/sources/t3dotgg"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,ai-agents,grok","timeRequired":"PT25M","video":{"@type":"VideoObject","name":"xAI just caught up (Grok 4.6 is here)","description":"Grok 4.6 from xAI is reviewed as a post-training upgrade over Grok 4.5, focused on long-running agentic tasks, coding, and visual/design work. It scores 61 on...","thumbnailUrl":"https://i.ytimg.com/vi/c7W8jpsjtCc/sddefault.jpg","uploadDate":"2026-08-13T11:41:45.674Z","duration":"PT25M","url":"https://api.daily.dev/r/PGnjC1Ych","embedUrl":"https://www.youtube.com/embed/c7W8jpsjtCc"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Theo - t3․gg","item":"https://daily.dev/sources/t3dotgg"},{"@type":"ListItem","position":3,"name":"xAI just caught up (Grok 4.6 is here)"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/xai-just-caught-up-grok-4-6-is-here--pgnjc1ych#faq","mainEntity":[{"@type":"Question","name":"What is different about Grok 4.6 compared to Grok 4.5?","acceptedAnswer":{"@type":"Answer","text":"Grok 4.6 is a new post-training pass on the same Grok 4.5 pre-trained base, not a new pretraining run, focused on long-running agentic tasks and visual/interactive work. It gained roughly five points on the Artificial Analysis Intelligence Index over 4.5, jumping benchmarks like Frontier Code from 56.6 to 61.3, but token usage per run rose over 30%, making it slower and more expensive per task than 4.5. Track how fast-moving model updates like this shift your agentic coding costs on daily.dev."}},{"@type":"Question","name":"How much does Grok 4.6 cost compared to Claude Opus 4.5 and GPT-5.1?","acceptedAnswer":{"@type":"Answer","text":"Grok 4.6 costs $2 per million input tokens and $6 per million output tokens, about 60% below Claude Opus 4.5's pricing and still cheaper than GPT-5.1. Real-world cost per task is around 84 cents, similar to Kimi K2, though this is more than double what Grok 4.5 cost due to lower token efficiency in the new version. Compare frontier model pricing shifts like this before picking a coding agent on daily.dev."}},{"@type":"Question","name":"Is Grok 4.6 good at 3D game development or design tasks?","acceptedAnswer":{"@type":"Answer","text":"No, Grok 4.6 performed poorly on 3D and design tasks in hands-on testing. It failed outright to render a 3D scene on a first attempt (producing a black screen), and after a fix attempt it shipped inverted controls and badly misplaced ground objects, ranking as the weakest 3D output tested among current models; design mockups were also judged generic and less polished than Claude or GPT-5.1 outputs. Weigh model-specific weaknesses like this when choosing an agent for UI or game work via daily.dev."}}]} ```