TL;DR
Anthropic’s Claude Fable 5 is priced at $10 per million input tokens and $50 per million output tokens, double Opus 4.8. Artificial Analysis data cited in the source material shows a 5.7% Intelligence Index gain, while public productivity proof remains limited to one unaudited customer story.
Anthropic’s Claude Fable 5 is priced at $10 per million input tokens and $50 per million output tokens, double the listed price of Opus 4.8, while independent benchmark data cited in the source material shows only a 5.7% aggregate Intelligence Index gain over the cheaper model.
The price is the firmest part of the story. According to Anthropic’s launch post, its platform pricing documents and the Fable product page, Fable 5 costs 2.0 times Opus 4.8 and about 3.3 times Sonnet 4.6 on list pricing.
The effective cost can fall with discounts. The source material says prompt caching reduces cache reads to $1 per million tokens, batch use is listed at $5 input and $25 output, and a blended 7:2:1 cache-hit-to-input-to-output mix produces a $7.70 per million token rate.
The benchmark gap is smaller than the price gap. Artificial Analysis scored Fable 5 at 64.9 on its Intelligence Index versus 61.4 for Opus 4.8, and GDPval-AA knowledge-work Elo at 1,932 versus 1,890. The same source material estimates a full Intelligence Index run at about $9,940 for Fable 5, compared with $4,970 for Opus 4.8.
Enterprise Buyers Face a Price Test
The numbers matter because AI budgets for 2026 are increasingly tied to measurable work output, not model prestige. If a company pays twice the model price for a 5.7% aggregate benchmark gain, it needs evidence that its own workload sits where Fable 5 performs better.
The data does not show that Fable 5 has no value. The source material says it is the highest-scoring public Claude model on aggregators and set records in five of 10 sub-benchmarks. But the value case changes if the buyer’s tasks fall outside those stronger areas, since the average gain is modest against the price premium.
MLOps for Decision Makers: The First Plain-Language Guide to Machine Learning Operations — Why 95% of AI Projects Fail and How to Make Sure Yours Doesn't (The AI Literacy Series Book 4)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From June Launch to Market
Fable 5 was launched on June 9, 2026 and moved toward general availability around July 1, according to the timeline in the source material. It sits above Opus 4.8 and Sonnet 4.6 in Anthropic’s active Claude lineup by price.
The source material also says Fable 5 and Mythos 5 use the same underlying weights and carry identical pricing, with the difference described as safeguards. That means reports grouping them together on price are treating them as the same priced model family, not mixing separate products.
“A codebase-wide migration was completed in a day and would otherwise have taken a whole team over two months by hand.”
— Anthropic launch post, citing Stripe
Leadership Transformation for the AI Era: Designing the Vision, Not Just Directing It. The 6×6 Blueprint for Strategic Trust and Decision … series for enterprise transformation)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Productivity Proof Remains Thin
The biggest gap is not pricing or benchmark scoring; it is workplace productivity evidence. The source material says there are zero controlled human-baseline studies and only one public enterprise proof point, the Stripe migration story credited in Anthropic’s launch post.
It is not yet clear whether that example reflects repeatable enterprise gains, a strong fit for one large Ruby codebase, or a result that depends on setup details not disclosed publicly. Claims that treat the Stripe example as independent proof or a measured human baseline are not established by the public record described in the source material.
End-to-End AI Evaluation: Building Effective Metrics, Pipelines, and Monitoring for LLM Systems
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Pilots Shift to Real Workloads
The next test is likely to come from enterprise pilots, not launch materials. Buyers comparing Fable 5 with Opus 4.8, Sonnet-tier models and rival frontier systems will need task-level benchmarks tied to their own codebases, documents, support queues or research workflows.
Anthropic could strengthen the case by publishing audited productivity studies, broader human-baseline comparisons or more detailed customer results. Until then, the confirmed story is clear: the price is documented, the benchmark gain is measurable, and the broader productivity claim is still developing.
AI prompt caching solutions
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
How much does Claude Fable 5 cost?
Claude Fable 5 is listed at $10 per million input tokens and $50 per million output tokens. That is twice Opus 4.8 on the prices cited in the source material.
Does Fable 5 perform better than Opus 4.8?
Yes, on the cited aggregate benchmark. Artificial Analysis scored Fable 5 at 64.9 versus 61.4 for Opus 4.8, a 5.7% gain.
Is the higher price justified?
That depends on the workload. The public data shows higher benchmark scores, but the source material says controlled productivity studies against a human baseline are not yet public.
What is the strongest productivity claim?
The strongest public claim is the Stripe migration story cited by Anthropic, where Fable 5 reportedly completed work in one day that would have taken a team more than two months by hand.
What should buyers watch next?
Buyers should watch for audited enterprise studies, task-specific benchmark results and evidence showing whether Fable 5’s gains hold across real company workflows.
Source: Thorsten Meyer AI
