--- title: My New Favorite Model | daily.dev description: A deep, personal review of Anthropic's Claude Fable 5.1 (and its unrestricted counterpart Mythos 5.1), covering pricing changes, benchmark results, data... canonical: https://daily.dev/posts/my-new-favorite-model-qle3aokdm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: My New Favorite Model | daily.dev og:description: A deep, personal review of Anthropic's Claude Fable 5.1 (and its unrestricted counterpart Mythos 5.1), covering pricing changes, benchmark results, data... og:url: https://daily.dev/posts/my-new-favorite-model-qle3aokdm og:image: https://api.daily.dev/og/posts/Qle3aOkdM.png og:image:alt: My New Favorite 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. # My New Favorite Model **[Theo - t3․gg](https://daily.dev/sources/t3dotgg)** · 55 min read · 1 upvotes · 0 comments ## Summary A deep, personal review of Anthropic's Claude Fable 5.1 (and its unrestricted counterpart Mythos 5.1), covering pricing changes, benchmark results, data retention policy shifts, anti-distillation history restrictions, and extensive real-world agentic coding use. The author landed dozens of PRs across two codebases using the model, walks through prompt-cache economics with real cost breakdowns, compares performance against Claude Opus 5 and GPT/Grok competitors, and details new behavioral quirks like reduced progress updates and denser writing style. Two sponsor segments for Blacksmith CI and General Translation are embedded mid-video. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How much cheaper are cache reads with Claude Fable 5.1 compared to Fable 5? Cache read pricing dropped 75%, from $1 per million cached input tokens down to 25 cents per million tokens. Base pricing stays the same at $10 per million input tokens and $50 per million output tokens. For highly agentic workloads that rely heavily on cache reads, overall savings can reach up to 45%, though typical workloads see roughly 25% savings, and cache write costs remain relatively high. _Teams weighing agentic coding costs can track how model pricing changes like this play out on daily.dev._ ### Why does Claude Fable 5.1 sometimes cost more per task despite cheaper cache pricing? Fable 5.1 can cost more per task because it generates roughly 1.7x more output tokens than Fable 5, according to artificial analysis benchmarking. On their intelligence index task, Fable 5.1 Max cost $3.76, about 20% more than Fable 5, even after the 75% cut to cache read pricing, since output token volume outweighs the cache savings on non-agentic single-response tasks. _Anyone budgeting for a model upgrade can weigh these token-cost tradeoffs before switching, tracked on daily.dev._ ### What changed with Claude's anti-distillation restrictions in Fable 5.1? New Claude accounts can no longer edit prior context within a multi-turn conversation; doing so now breaks the thread entirely instead of just invalidating the cache as before. This closes a workaround people used to extract hidden reasoning tokens from earlier turns. Anthropic plans to roll this restriction out gradually to all accounts, which will require developers building on the Claude API to only append to chat history going forward. _Developers building on Claude's API can stay ahead of history-handling breakages tracked on daily.dev._ --- Tags: [#ai](https://daily.dev/tags/ai), [#ai-agents](https://daily.dev/tags/ai-agents), [#claude](https://daily.dev/tags/claude), [#anthropic](https://daily.dev/tags/anthropic), [#ai-inference](https://daily.dev/tags/ai-inference) [View this post on daily.dev](https://daily.dev/posts/my-new-favorite-model-qle3aokdm) ```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":"My New Favorite Model","url":"https://daily.dev/posts/my-new-favorite-model-qle3aokdm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/my-new-favorite-model-qle3aokdm"},"datePublished":"2026-09-03T08:41:50.644Z","dateModified":"2026-09-13T22:01:19.746Z","description":"A deep, personal review of Anthropic's Claude Fable 5.1 (and its unrestricted counterpart Mythos 5.1), covering pricing changes, benchmark results, data...","image":"https://i.ytimg.com/vi/r_dw-1109Ag/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/r_dw-1109Ag/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/my-new-favorite-model-qle3aokdm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,ai-agents,claude,anthropic,ai-inference","timeRequired":"PT55M","video":{"@type":"VideoObject","name":"My New Favorite Model","description":"A deep, personal review of Anthropic's Claude Fable 5.1 (and its unrestricted counterpart Mythos 5.1), covering pricing changes, benchmark results, data...","thumbnailUrl":"https://i.ytimg.com/vi/r_dw-1109Ag/sddefault.jpg","uploadDate":"2026-09-03T08:41:50.644Z","duration":"PT55M","url":"https://api.daily.dev/r/Qle3aOkdM","embedUrl":"https://www.youtube.com/embed/r_dw-1109Ag"}} {"@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":"My New Favorite Model"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/my-new-favorite-model-qle3aokdm#faq","mainEntity":[{"@type":"Question","name":"How much cheaper are cache reads with Claude Fable 5.1 compared to Fable 5?","acceptedAnswer":{"@type":"Answer","text":"Cache read pricing dropped 75%, from $1 per million cached input tokens down to 25 cents per million tokens. Base pricing stays the same at $10 per million input tokens and $50 per million output tokens. For highly agentic workloads that rely heavily on cache reads, overall savings can reach up to 45%, though typical workloads see roughly 25% savings, and cache write costs remain relatively high. Teams weighing agentic coding costs can track how model pricing changes like this play out on daily.dev."}},{"@type":"Question","name":"Why does Claude Fable 5.1 sometimes cost more per task despite cheaper cache pricing?","acceptedAnswer":{"@type":"Answer","text":"Fable 5.1 can cost more per task because it generates roughly 1.7x more output tokens than Fable 5, according to artificial analysis benchmarking. On their intelligence index task, Fable 5.1 Max cost $3.76, about 20% more than Fable 5, even after the 75% cut to cache read pricing, since output token volume outweighs the cache savings on non-agentic single-response tasks. Anyone budgeting for a model upgrade can weigh these token-cost tradeoffs before switching, tracked on daily.dev."}},{"@type":"Question","name":"What changed with Claude's anti-distillation restrictions in Fable 5.1?","acceptedAnswer":{"@type":"Answer","text":"New Claude accounts can no longer edit prior context within a multi-turn conversation; doing so now breaks the thread entirely instead of just invalidating the cache as before. This closes a workaround people used to extract hidden reasoning tokens from earlier turns. Anthropic plans to roll this restriction out gradually to all accounts, which will require developers building on the Claude API to only append to chat history going forward. Developers building on Claude's API can stay ahead of history-handling breakages tracked on daily.dev."}}]} ```