Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d270Policy30d65, 13 pendingSources1,015

Claude Fable 5

AnthropicReleased Jun 9, 2026Current

Score basis

Claude Fable 5 is at the frontier in coding and agentic tasks; strong in reasoning, factuality, and long context; and capable in math, multimodal tasks, and instruction following. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Frontier2 capabilities
  1. CodingLeads9 of 10
  2. Agentic−3.7%7 of 7
Strong3 capabilities
  1. Reasoning−3.4%6 of 6
  2. Factuality−4.4%3 of 4
  3. Long Context−7.9%1 of 3
Capable3 capabilities
  1. Math−12.0%5 of 5
  2. Multimodal−16.4%2 of 6
  3. Instruction Following−20.3%2 of 3
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$10.00input$50.00outputper million tokens

From Anthropic's own price page · 5 providers tracked · All prices

Costs more than 95% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

155results on97benchmarks

  • 31 independently verified
  • 18 aggregator
  • 48 vendor-reported
  • 58 cross-referenced

From 32 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

42 papers reference Claude Fable 5

Claude Fable 5 benchmark results

155 results on 97 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Coding

FrontierFull coding ranking

Leads the field9 of 10 ranked benchmarks measured

Show 8 more coding resultsHide 8 coding results

Agentic

FrontierFull agentic ranking

3.7% behind the leader7 of 7 ranked benchmarks measured

Show 7 more agentic resultsHide 7 agentic results

Reasoning

StrongFull reasoning ranking

3.4% behind the leader6 of 6 ranked benchmarks measured

Show 12 more reasoning resultsHide 12 reasoning results

Factuality

StrongFull factuality ranking

4.4% behind the leader3 of 4 ranked benchmarks measured

Show 1 more factuality resultHide 1 factuality result

Long Context

StrongFull long context ranking

7.9% behind the leader1 of 3 ranked benchmarks measured

Math

CapableFull math ranking

12.0% behind the leader5 of 5 ranked benchmarks measured

Multimodal

CapableFull multimodal ranking

16.4% behind the leader2 of 6 ranked benchmarks measured

Show 4 more multimodal resultsHide 4 multimodal results

Instruction Following

CapableFull instruction following ranking

20.3% behind the leader2 of 3 ranked benchmarks measured

Show 1 more instruction following resultHide 1 instruction following result

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 84 more resultsHide 84 results

Claude Fable 5: common questions

Who makes Claude Fable 5?

Claude Fable 5 is made by Anthropic.

When was Claude Fable 5 released?

Claude Fable 5 was released on Jun 9, 2026, according to Artificial Analysis.

What is Claude Fable 5 good at?

Claude Fable 5 is at the frontier in coding and agentic tasks; strong in reasoning, factuality, and long context; and capable in math, multimodal tasks, and instruction following. Too few results yet to rate safety or multilingual tasks.

How much does Claude Fable 5 cost?

Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens, according to Anthropic's own price page. We track its price at 5 providers. At a mix of three input tokens to one output token, it costs more than 95% of the 330 priced models we track.

How many benchmarks has Claude Fable 5 been tested on?

We track 155 results for Claude Fable 5 on 97 benchmarks from 32 sources, 31 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does Claude Fable 5 support?

OpenRouter lists tool calling, structured outputs, and reasoning for Claude Fable 5.

About this record

Where Claude Fable 5's numbers come from, and every name it appears under.

Tracked since
Jun 10, 2026
Newest source mention
Sep 18, 2026

Where the results come from

Verification: 155 scores · 31 independently verified · 18 aggregator-attributed · 58 vendor cross-reference · 48 vendor-reported. How these tiers are assigned

From 32 sources on 13 sites. Hugging Face supplies 54 of them; the 31 independently verified results come from 8 sites. Bars are coloured by trust tier.

  • huggingface.co54
  • www-cdn.anthropic.com29
  • artificialanalysis.ai21
  • api.llm-stats.com13
  • arcprize.org9
  • livebench.ai7
  • anthropic.com6
  • datasets-server.huggingface.co4
  • x.ai4
  • epoch.ai3
  • labs.scale.com2
  • lmarena.ai2
  • simple-bench.com1

Also known as

How our sources name Claude Fable 5 at each reasoning setting.

SettingShort formLong formAPI id
lowclaude fable 5 (low)——
mediumclaude fable 5 (medium)——
highClaude Fable 5 (Extra High) claude fable 5 (high)—claude-fable-5-high
xhighclaude fable 5 (xhigh)——
maxclaude fable 5 (max)claude fable 5 (adaptive reasoning, max effort, opus 4.8 fallback)—
Also listed asanthropic claude fable 5claude 5 fableClaude Fclaude fableclaude fable5claude-fable-5:batchclaude-fable-latestfable 5