--- title: Fable 5.1 (Fully Tested & Real cost comparisons): It's A... description: Anthropic released Claude Fable 5.1 (alongside an unrestricted Mythos 5.1 variant) on September 1st, keeping input/output token pricing unchanged from Fable 5... canonical: https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE! | daily.dev og:description: Anthropic released Claude Fable 5.1 (alongside an unrestricted Mythos 5.1 variant) on September 1st, keeping input/output token pricing unchanged from Fable 5... og:url: https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc og:image: https://api.daily.dev/og/posts/qYSrGZZqc.png og:image:alt: Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE! 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. # Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE! **[AICodeKing](https://daily.dev/sources/aicodeking)** · 16 min read · 0 upvotes · 0 comments ## Summary Anthropic released Claude Fable 5.1 (alongside an unrestricted Mythos 5.1 variant) on September 1st, keeping input/output token pricing unchanged from Fable 5 but cutting cache read prices by 75% (from $1 to $0.25 per million). The model shows large benchmark jumps over Fable 5, especially on terminal and wristwatch-style agentic coding tests, scoring 92.5% on the reviewer's own benchmark, the highest ever recorded there. Three breaking API changes ship: forced tool use is removed, thinking blocks are model-locked, and editing earlier turns invalidates thinking blocks (explicitly framed by Anthropic as anti-distillation measures). Real-world cost breakdowns show cache writes, which didn't get cheaper, now dominate bills (up to 66% of cost despite being under 10% of tokens). Community reports also flag inconsistent refusal behavior, with an estimated 80% of certain security-related prompts silently routed to Opus, plus denser, less readable writing style changes documented in Anthropic's own migration guide. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What changed in Claude Fable 5.1 API pricing compared to Fable 5? Input and output token pricing stayed the same at $10 per million input and $50 per million output tokens, but cache read pricing dropped 75%, from $1 to $0.25 per million tokens. Cache write pricing did not change, remaining at $1.25 per million for the five-minute cache and $2.00 per million for the one-hour cache. Anthropic says this yields roughly 25% savings on typical workloads and up to 45% on highly agentic long-running sessions. _Track pricing shifts like this cache-read cut before they blow up your Claude API bill, on daily.dev._ ### What breaking changes does Claude Fable 5.1 introduce for API users? Three breaking changes ship with Fable 5.1: forced tool use (tool_choice set to any or a specific tool name) now returns a 400 error because it would bypass always-on thinking; thinking blocks generated by Fable 5.1 can only be read by Fable 5.1 and Mythos 5.1, so swapping to an older model mid-conversation silently drops them; and for accounts created after August 31st, editing any earlier turn, the system prompt, or the tools array invalidates every thinking block after it, making conversations effectively append-only. _Developers upgrading Claude integrations check breaking API changes like these on daily.dev before shipping._ ### Why are Claude Fable 5.1 API bills now dominated by cache writes instead of cache reads? Because cache read prices dropped 75% while cache write prices stayed flat, cache writes now make up the majority of typical bills despite being a small share of total tokens. One real-world breakdown showed cache writes at 9.4% of tokens but 65.8% of a $1,023.78 bill, while cache reads were 89.8% of tokens but only 12.6% of cost, since the price cut only benefits cache reads, not writes. _Anyone budgeting agentic API usage weighs cost breakdowns like this on daily.dev before scaling workloads._ --- 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/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc) ```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":"Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE!","url":"https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc"},"datePublished":"2026-09-02T12:52:45.785Z","dateModified":"2026-09-02T12:53:49.522Z","description":"Anthropic released Claude Fable 5.1 (alongside an unrestricted Mythos 5.1 variant) on September 1st, keeping input/output token pricing unchanged from Fable 5...","image":"https://i.ytimg.com/vi/UZ2PRAjEPRY/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/UZ2PRAjEPRY/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AICodeKing","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":"AICodeKing","logo":"https://media.daily.dev/image/upload/s--x7nDUfWj--/f_auto,q_auto/v1768208312/logos/aicodeking","url":"https://daily.dev/sources/aicodeking"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,ai-coding,claude,anthropic","timeRequired":"PT16M","video":{"@type":"VideoObject","name":"Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE!","description":"Anthropic released Claude Fable 5.1 (alongside an unrestricted Mythos 5.1 variant) on September 1st, keeping input/output token pricing unchanged from Fable 5...","thumbnailUrl":"https://i.ytimg.com/vi/UZ2PRAjEPRY/sddefault.jpg","uploadDate":"2026-09-02T12:52:45.785Z","duration":"PT16M","url":"https://api.daily.dev/r/qYSrGZZqc","embedUrl":"https://www.youtube.com/embed/UZ2PRAjEPRY"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AICodeKing","item":"https://daily.dev/sources/aicodeking"},{"@type":"ListItem","position":3,"name":"Fable 5.1 (Fully Tested & Real cost comparisons): It's A GREAT Model but there's still ONE ISSUE!"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/fable-5-1-fully-tested-real-cost-comparisons-it-s-a-great-model-but-there-s-still-one-issue--qysrgzzqc#faq","mainEntity":[{"@type":"Question","name":"What changed in Claude Fable 5.1 API pricing compared to Fable 5?","acceptedAnswer":{"@type":"Answer","text":"Input and output token pricing stayed the same at $10 per million input and $50 per million output tokens, but cache read pricing dropped 75%, from $1 to $0.25 per million tokens. Cache write pricing did not change, remaining at $1.25 per million for the five-minute cache and $2.00 per million for the one-hour cache. Anthropic says this yields roughly 25% savings on typical workloads and up to 45% on highly agentic long-running sessions. Track pricing shifts like this cache-read cut before they blow up your Claude API bill, on daily.dev."}},{"@type":"Question","name":"What breaking changes does Claude Fable 5.1 introduce for API users?","acceptedAnswer":{"@type":"Answer","text":"Three breaking changes ship with Fable 5.1: forced tool use (tool_choice set to any or a specific tool name) now returns a 400 error because it would bypass always-on thinking; thinking blocks generated by Fable 5.1 can only be read by Fable 5.1 and Mythos 5.1, so swapping to an older model mid-conversation silently drops them; and for accounts created after August 31st, editing any earlier turn, the system prompt, or the tools array invalidates every thinking block after it, making conversations effectively append-only. Developers upgrading Claude integrations check breaking API changes like these on daily.dev before shipping."}},{"@type":"Question","name":"Why are Claude Fable 5.1 API bills now dominated by cache writes instead of cache reads?","acceptedAnswer":{"@type":"Answer","text":"Because cache read prices dropped 75% while cache write prices stayed flat, cache writes now make up the majority of typical bills despite being a small share of total tokens. One real-world breakdown showed cache writes at 9.4% of tokens but 65.8% of a $1,023.78 bill, while cache reads were 89.8% of tokens but only 12.6% of cost, since the price cut only benefits cache reads, not writes. Anyone budgeting agentic API usage weighs cost breakdowns like this on daily.dev before scaling workloads."}}]} ```