What is Claude Fable 5.1?
Fable 5.1 is the second Fable model and the most capable one Anthropic sells to everyone. It reads text and images, writes text, and works across a 1M-token context with up to 128K tokens of output. Thinking is always on and defaults to high effort on the API, and its knowledge stops at June 2026. Mythos 5.1 is the same model with looser safeguards, offered only to vetted partners.
It is current, but no longer the default. Anthropic now tells developers to start with Opus 5.5, which beat Fable 5.1 on every test in its own launch table at 40% of the token price, and to step up only when Opus 5.5 still falls short.
What changed from Fable 5?
Mostly cost and capability: the same token prices with far cheaper cache reads, better scores across Anthropic's tests, and a few breaking API changes.
- Price
- Input and output stay at $10 / $50, but cache reads drop from $1 to $0.25 per million tokens.
- Capability
- Terminal-Bench-Science 0.1 more than doubled, from 24.7% to 52.6%, and AutomationBench rose from 17.1% to 31.4%.
- Safeguards
- Anthropic says Claude Code users see about 60% fewer cyber interventions per session, and it may now help find vulnerabilities, though not write exploits.
- API
- Forced tool use returns an error, and editing earlier turns invalidates its thinking blocks.
Who is Claude Fable 5.1 best for?
Claude Fable 5.1 is best for the hardest coding and analysis problems that Opus 5.5 can't solve, agent jobs that run for hours, document-heavy analysis, and Claude Max or Team premium users who already have it included.
Problems Opus 5.5 can't crack
Anthropic and many users treat it as the step-up model for multi-file bugs, architecture calls and audits where Opus 5.5 at high effort still misses.
Agents that run for hours
It plans multi-step jobs across many apps, checks its own work and recovers when a step fails, whether in Claude Code, Cowork or as a managed agent.
Document-heavy analysis
It reads dense charts, filings and tables inside PDFs and spreadsheets, which suits finance, legal and analytics work.
Claude Max and Team premium users
It is included up to half of the weekly usage limit on Max (from $100 a month) and Team premium seats, so there is no extra bill until that cap.
What's included with Claude Fable 5.1?
Five effort levels, cheap cache reads, half-price batch jobs, automatic fallback to Opus, progress updates between tool calls, and every major cloud.
- Effort levels
- Five settings from low to max, defaulting to high, and a beta lets you change effort mid-conversation without losing the prompt cache.
- Cheap cache reads
- Cached input costs $0.25 per million tokens, a quarter of Fable 5's rate, and Anthropic says cache reads make up most of the cost in context-heavy agent work.
- Batch API
- Half price for jobs that can wait: $5 per million input tokens and $25 per million output tokens.
- Automatic fallback
- Requests its safeguards block can re-run on Opus 4.8 or Opus 5, by default in the Claude apps and through a fallback setting on the API.
- Progress updates
- A beta option returns its short status notes between tool calls, so long agent runs don't look silent to your users.
- Runs on every major cloud
- Available on the Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry and Claude Platform on AWS; US-only inference costs 1.1x.
How much does Claude Fable 5.1 cost?
Claude Fable 5.1 costs $10 per million input tokens and $50 per million output tokens on the API; Max (from $100 a month) includes it up to half your weekly limit, while Pro ($20 a month) charges it as usage credits.
- Cache reads $0.25, cache writes $12.50 (5-minute) or $20 (1-hour) per million tokens
- Batch API halves it to $5 / $25
- Also on Amazon Bedrock, Google Cloud and Microsoft Foundry
- Fable 5.1 runs only on pay-as-you-go usage credits
- Usage credits are billed at standard API rates
- Opus, Sonnet and Haiku stay within the plan's limits
- Fable 5.1 included, up to 50% of your weekly usage limit
- 5x Pro usage for $100, or 20x for $200
- Usage credits keep Fable running past that cap
- Fable included up to 50% of each member's weekly limit
- Standard seats ($25) use Fable only through usage credits
- For teams of 2 to 150
The Free plan has no Fable access. Paying yearly brings Pro to $17 a month ($200 up front) and Team seats to $20 (standard) or $100 (premium) a month; Max is billed monthly only. Enterprise costs $20 per seat per month, billed yearly, plus usage at API rates. US-only inference costs 1.1x the input and output price. When the safeguards hand a request to an Opus model, that answer is billed at Opus rates, but a refusal in the biology, distillation or frontier LLM categories is billed even before any output. Fable needs 30-day data retention by default. Full pricing on claude.com. We checked these prices on .
How do readers rate Claude Fable 5.1?
No one has rated Claude Fable 5.1 yet. Be the first.
Sign in to rate. One rating per account.
Claude Fable 5.1 pros and cons
It is Anthropic's strongest model for hard, long jobs and cheaper to run than Fable 5, but safety reroutes, fast-draining limits and a cheaper Opus 5.5 that beats it on Anthropic's own tests weigh against it.
Pros
- Many Reddit and Hacker News users keep it as the escalation model for multi-file bugs, architecture decisions and audits, where they say it catches what Opus misses.
- Several early users say it reads more plainly than Fable 5 and Opus 5, though a few still find its prose dense.
- Cache reads cost 75% less than on Fable 5, which Anthropic says cuts typical costs by about 25% and cache-heavy agent work by up to 45%.
- Far ahead of Fable 5 on Anthropic's tests: 52.6% on Terminal-Bench-Science 0.1 against 24.7%, and 55.8% on Terminal-Bench 4.0 against 42.0%.
Cons
- Safety classifiers still hand many requests to Opus 4.8 or Opus 5, and users report reroutes on ordinary auth reviews, firmware work and even the word "biological" in a game file.
- It drains plan limits fast: Max and Team premium users on Reddit and Hacker News report hitting five-hour or weekly limits within hours or days, and on Pro it runs only on paid usage credits.
- It costs 2.5 times Opus 5.5 per token yet trails it on every test in Anthropic's own table, and two Hacker News testers found Opus 5.5 matched it on code review, one at under a quarter of the cost.
- Code written for Fable 5 can break: forced tool use returns an error, editing earlier turns invalidates thinking blocks, and it needs 30-day data retention unless Anthropic approves otherwise.
What are the best Claude Fable 5.1 alternatives?
GPT-6 Astra at the same price from OpenAI, Sonnet 5 for speed at a fifth of the cost, and Opus 4.8 for verified security work.
GPT-6 AstraPaidChoose it when you want OpenAI's top model at the same $10 / $50 list price; it led Fable 5.1 on Terminal-Bench 4.0 and AutomationBench in Anthropic's own comparison, though cached input costs $1 against Fable's $0.25.
Claude Sonnet 5FreeChoose it when speed and cost matter more than peak reasoning; it costs $2 / $10 per million tokens, a fifth of Fable 5.1, with the same 1M context, but its knowledge stops at January 2026.
Claude Opus 4.8PaidChoose Claude Opus 4.8 for defensive security work: Fable 5.1 sends offensive cyber requests to Opus 4.8 anyway, and Opus 4.8 is covered by Anthropic's Cyber Verification Program at $5 / $25 per million tokens.
Questions about Claude Fable 5.1
What is the difference between Claude Fable 5.1 and Claude Mythos 5.1?
They are the same model at the same price with different safeguards. Fable 5.1 is sold to everyone and blocks or reroutes some cybersecurity and biology work, while Mythos 5.1 has looser limits and is offered only to vetted partners through Anthropic's trusted access programs.
Who is Claude Fable 5.1 for?
Claude Fable 5.1 is best for the hardest coding and analysis problems that Opus 5.5 can't solve, agent jobs that run for hours, document-heavy analysis, and Claude Max or Team premium users who already have it included.
Is Claude Fable 5.1 free?
No. Claude's Free plan has no Fable access. Pro and Team standard seats can use it only through usage credits billed at API rates, while Max plans and Team premium seats include it up to half of the weekly usage limit.
How much does Claude Fable 5.1 cost?
Claude Fable 5.1's paid plans are Claude API at $10 / $50 per million input / output tokens, Pro at $20 per month, Max at From $100 per month and Team premium seat at $125 per seat per month.
What are the best Claude Fable 5.1 alternatives?
The closest alternatives on Recaply are GPT-6 Astra, Claude Sonnet 5 and Claude Opus 4.8.
Where can I try Claude Fable 5.1?
On its official website, anthropic.com.
Claude Fable 5.1 vs Opus 5.5: which should I use?
Start with Opus 5.5. It costs 40% of Fable 5.1's token price and scored higher on every test in Anthropic's launch table. Anthropic recommends moving up to Fable 5.1 only when your own tests on Opus 5.5 at higher effort still fall short.
Why did my Claude Fable 5.1 conversation switch to Opus?
A safety classifier flagged something in the conversation, including files, memory or search results, and Claude re-ran the request on Opus 4.8 for offensive security or Opus 5 for biology. You can turn automatic switching off in Settings, then Capabilities.
Can I use Claude Fable 5.1 with zero data retention?
Only with Anthropic's approval. Fable 5.1 keeps data for 30 days for safety monitoring by default. Eligible enterprise customers can use it with zero data retention until Enterprise Frontier Safeguards, which store data in the customer's own cloud, become available.
Our verdict
Fable 5.1 is still Anthropic's strongest general model, but for most people it is now the step-up model rather than the daily one.
Opus 5.5 beat it on every test in Anthropic's own launch table at 40% of the price, and many users now keep Fable for the multi-file bugs, audits and plans where Opus 5.5 still misses.
- Skip it for
- Security or biology work, where its safeguards reroute requests, and Pro users who would pay for every token of it.
- Check first
- Run your hardest tasks on Opus 5.5 at high effort first, and pay for Fable 5.1 only where it clearly does better.
Claude Fable 5
Claude Opus 4.5
Claude Opus 4.7
Claude Sonnet 4.6
Muse Spark 1.3