--- title: Opus 5 is my new go-to model | daily.dev description: Claude Opus 5 has just launched and is being praised as a strong everyday coding model. It outperforms Claude Sonnet on most benchmarks and comes close to... canonical: https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Opus 5 is my new go-to model | daily.dev og:description: Claude Opus 5 has just launched and is being praised as a strong everyday coding model. It outperforms Claude Sonnet on most benchmarks and comes close to... og:url: https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p og:image: https://api.daily.dev/og/posts/kQ8PhQP7P.png og:image:alt: Opus 5 is my new go-to model 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. # Opus 5 is my new go-to model **[Theo - t3․gg](https://daily.dev/sources/t3dotgg)** · 44 min read · 1 upvotes · 0 comments ## Summary Claude Opus 5 has just launched and is being praised as a strong everyday coding model. It outperforms Claude Sonnet on most benchmarks and comes close to Claude's flagship model (referred to as 'Fable/Mythos') at roughly half the token price, though real-world cost savings are closer to 20-25% due to higher token usage per task. Key advantages include zero data retention (enabling enterprise use cases), full usage limits on subscription plans (vs. 50% for the flagship), better instruction-following behavior, and a balance between the thoroughness of OpenAI-style models and the taste/quality of Anthropic's top models. It is Anthropic's most aligned model yet, with dual-use capabilities deliberately reduced. The author recommends it as the new default for most coding tasks, especially for Anthropic subscribers burning through limits with the flagship model. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the token pricing for Claude Opus 5 compared to Claude Opus 4.8 and other frontier models? Claude Opus 5 costs $5 per million input tokens and $25 per million output tokens, the same price as its predecessor Opus 4.8 but with much improved performance. This makes it roughly half the sticker price of a comparable rival model priced at $10 per million input and $50 per million output, though real-world token usage narrows the effective savings to about 20-25% rather than a full 50%. _Track pricing and performance shifts across frontier coding models as new releases land, right on daily.dev._ ### Is Claude Opus 5 more token-efficient than Claude Opus 4.8 or its main rival for coding tasks? Claude Opus 5 is more token-efficient than Opus 4.8 but still less efficient than its main rival, using around 37,000 tokens per task on average versus roughly 33,000 for the rival, according to artificial analysis benchmarks. This means Opus 5 takes longer to respond, burns through context windows faster, and delivers a smaller real-world cost advantage than its per-token pricing suggests. _Compare real-world token efficiency, not just sticker price, before picking a coding model on daily.dev._ ### How does Claude Opus 5 perform on agentic coding benchmarks like Frontier Code and ARC-AGI-3? Claude Opus 5 set a new high score of 43.3% on Frontier Code, a benchmark measuring code maintainability and mergeability, up from a previous best of 34.4%. It also scored around 30% on the ARC-AGI-3 reasoning benchmark, a dramatic jump from near 1% shortly after that benchmark's release, and performed comparably to a rival model on agentic browsing tasks (90.8 vs 90.4). _Keep up with fast-moving coding benchmark results to judge which model actually fits your workflow, via daily.dev._ ## Similar posts on daily.dev - [Anthropic’s Opus 5 is almost Fable 5](https://daily.dev/posts/anthropic-s-opus-5-is-almost-fable-5-vlen80a1c) · The New Stack · 9 upvotes · 2 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#ai-coding](https://daily.dev/tags/ai-coding), [#claude](https://daily.dev/tags/claude), [#anthropic](https://daily.dev/tags/anthropic) [View this post on daily.dev](https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p) ```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":"Opus 5 is my new go-to model","url":"https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p"},"datePublished":"2026-07-25T10:21:01.204Z","dateModified":"2026-09-14T07:27:40.595Z","description":"Claude Opus 5 has just launched and is being praised as a strong everyday coding model. It outperforms Claude Sonnet on most benchmarks and comes close to...","image":"https://i.ytimg.com/vi/cIgoqAy_Vs8/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/cIgoqAy_Vs8/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/opus-5-is-my-new-go-to-model-kq8phqp7p","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,ai-coding,claude,anthropic","timeRequired":"PT44M","video":{"@type":"VideoObject","name":"Opus 5 is my new go-to model","description":"Claude Opus 5 has just launched and is being praised as a strong everyday coding model. It outperforms Claude Sonnet on most benchmarks and comes close to...","thumbnailUrl":"https://i.ytimg.com/vi/cIgoqAy_Vs8/sddefault.jpg","uploadDate":"2026-07-25T10:21:01.204Z","duration":"PT44M","url":"https://api.daily.dev/r/kQ8PhQP7P","embedUrl":"https://www.youtube.com/embed/cIgoqAy_Vs8"}} {"@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":"Opus 5 is my new go-to model"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/opus-5-is-my-new-go-to-model-kq8phqp7p#faq","mainEntity":[{"@type":"Question","name":"What is the token pricing for Claude Opus 5 compared to Claude Opus 4.8 and other frontier models?","acceptedAnswer":{"@type":"Answer","text":"Claude Opus 5 costs $5 per million input tokens and $25 per million output tokens, the same price as its predecessor Opus 4.8 but with much improved performance. This makes it roughly half the sticker price of a comparable rival model priced at $10 per million input and $50 per million output, though real-world token usage narrows the effective savings to about 20-25% rather than a full 50%. Track pricing and performance shifts across frontier coding models as new releases land, right on daily.dev."}},{"@type":"Question","name":"Is Claude Opus 5 more token-efficient than Claude Opus 4.8 or its main rival for coding tasks?","acceptedAnswer":{"@type":"Answer","text":"Claude Opus 5 is more token-efficient than Opus 4.8 but still less efficient than its main rival, using around 37,000 tokens per task on average versus roughly 33,000 for the rival, according to artificial analysis benchmarks. This means Opus 5 takes longer to respond, burns through context windows faster, and delivers a smaller real-world cost advantage than its per-token pricing suggests. Compare real-world token efficiency, not just sticker price, before picking a coding model on daily.dev."}},{"@type":"Question","name":"How does Claude Opus 5 perform on agentic coding benchmarks like Frontier Code and ARC-AGI-3?","acceptedAnswer":{"@type":"Answer","text":"Claude Opus 5 set a new high score of 43.3% on Frontier Code, a benchmark measuring code maintainability and mergeability, up from a previous best of 34.4%. It also scored around 30% on the ARC-AGI-3 reasoning benchmark, a dramatic jump from near 1% shortly after that benchmark's release, and performed comparably to a rival model on agentic browsing tasks (90.8 vs 90.4). Keep up with fast-moving coding benchmark results to judge which model actually fits your workflow, via daily.dev."}}]} ```