Claude Fable 5

Name: Claude Fable 5
Brand: anthropic
Price: 10.00 USD
Availability: OnlineOnly

Claude:

3 Haiku

3 Opus

3 Sonnet

Sonnet 3.6 (2022-10-22)

Fable 5

Anthropic's first public Mythos-class model, for autonomous knowledge work and coding. Text, image and file input, 1M token context. Built for long, ambiguous, multi-step tasks that used to need check-ins, self-correcting through verification loops. Falls back to Claude Opus 4.8 on high-risk queries.

Ranks#1Creative Coding#1Creative Writing#1Data Analysis#1Game Development#1Nostalgia#1Practical Tasks

Website OpenRouter Docs API Blog

Provider: Anthropic
Released: 2026-06-09
Size: XLARGE
Free API (OpenRouter): No
Per 1M tokens: $10 in · $50 out
Rival’s tab: $19.70; 58/58 receipts survived

Free API availability

No free endpoint for Claude Fable 5 was found in Rival’s OpenRouter listings. Other providers or chat apps may have separate free offers.

Provider data: 8 Sep 2026OpenRouter usage limits

View API options

Get API accessProvider and language code samples

4 hostsInOutContextUptime

Amazon Bedrock$10.00 in$50.00 out1M—Azure AI Foundry$10.00 in$50.00 out1M99.4% up Anthropic$10.00 in$50.00 out1M100% up Google Vertex AI$10.00 in$50.00 out1M98.9% up

Per million tokens. Prices and uptime via OpenRouter, checked 8 Sep 2026.

Provider

fromimport openai  OpenAI

client = OpenAI(
"https://openrouter.ai/api/v1"    base_url=,
"$OPENROUTER_API_KEY"    api_key=,
)

response = client.chat.completions.create(
"anthropic/claude-fable-5"    model=,
"role""user""content""Hello!"    messages=[{: , : }],
)
print(response.choices[0].message.content)

Set OPENROUTER_API_KEY with your OpenRouter API key from openrouter.ai/keys.

Also on Amazon Bedrock · Google Vertex AI · Microsoft Foundry

Deep analysisPsychometrics, taste index, writing DNA

Personality Analysis

The Long-Haul Closer

Class

Lawful Good

Pushes decisively toward the legally and ethically correct action and is willing to commit to a verdict on charged questions ("yes, justifiable") rather than retreat into both-sides hedging. Refuses false binaries and will not let the user rationalize a harmful path. Ships with hard safeguards that defer a small fraction of genuinely high-risk queries (cybersecurity, biology) to Claude Opus 4.8.

When you push back

Leads with a clear answer, then backs it with cited authority (21 CFR 314.80, the CBE-0 pathway, Rule 10b-5) rather than vague appeals to judgment. Corrects a flawed premise before solving it, holds two true things at once when a question is rhetorically loaded, and stays anchored in concrete, executable next steps. Will not be argued off a well-reasoned position under social pressure.

Tasting Notes

Commits to a VerdictCites Chapter and VerseReframes the PremiseMarathon StaminaGuardrails on the High-Risk Stuff

SubjectiveBench

Taste Index

Across 56 scored outputs

How this is measured →

600.60xCompetent

0100headroom →

Taste is judged on an uncapped scale, originality first. The space past 100 is craft today's models rarely reach.

Craft69

Originality69

Plays it safe

share of outputs that are the default answer

21%

Writing DNA

Stylometric Fingerprint

Based on 27 text responses

Tick = global average

Vocabulary Diversity60%

Unique words vs. total words. Higher = richer vocabulary.

Sentence Length22.3 words

Average words per sentence.

Hedging0.49

"Might", "perhaps", "arguably" per 100 words.

Bold Formatting4.0

**Bold** markers per 1,000 characters.

List Usage3.5

Bullet and numbered list items per 1,000 characters.

Section Structure1.09

Markdown headings per 1,000 characters.

Emoji Usage0.58

Emoji per 1,000 characters.

Transitions0.05

"However", "moreover", "furthermore" per 100 words.

Opening Habits

Starts with heading (74%)

Consistency

73%

Across 27 responses

Favorites

Movie

2001: A Space Odyssey

19688.0

Album

Kind of Blue

Miles Davis

Book

Gödel, Escher, Bach

Douglas R. Hofstadter

City

Istanbul

Turkey

Game

Outer Wilds

Indie, Adventure4.4

Model Responses

53 outputs · $19.68 tracked across 53 receipts

Keep exploring

COMPARE

Claude Fable 5 vs MiniMax M3

Real outputs compared side by side

RANKINGS

Best AI for Complex Reasoning

Sally's brothers, GPT-3 training FLOPs, a document editor syncing on client c...

Compare Claude Fable 5

Alternatives to Claude Fable 5

Claude Fable 5

Free API availability

The Long-Haul Closer

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Claude Fable 5 vs MiniMax M3

Best AI for Complex Reasoning

Compare Claude Fable 5

Alternatives to Claude Fable 5

Claude Fable 5

Free API availability

The Long-Haul Closer

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

Claude Fable 5 vs MiniMax M3

Best AI for Complex Reasoning