NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Claude Fable 5.2
Anthropic logo

Claude Fable 5.2

by Anthropic · Released Nov 2026

Multimodal1M Context
95.4
avg score
Rank #1
Compare
#1 ranked model today
Context
1.0M tokens (~500 books)
Input $/1M
$8.00
Output $/1M
$40.00
Type
multimodal
License
Proprietary
Benchmarks
10 tested
Data as of
About

Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible...

Tested on 10 benchmarks · BenchGecko score 95.4. Top scores: FrontierMath-Tiers-1-3-v2-Private (100.0%), FrontierMath-Tier-4-v2-Private (100.0%), OTIS Mock AIME 2024-2025 (99.7%).

Looking for similar performance at lower cost?
Step 3.5 Flash scores 89.5 (94% as good) at $0.10/1M input · 99% cheaper
Capabilities
math
99.9
#1 globally
knowledge
71.8
#15 globally
general
88.4
#1 globally
Benchmark Scores
Compare All
Tested on 10 benchmarks · Ranked across 3 categories
Score Distribution (all 326 models)
0255075100
▲ You are here
mathCompare math →
FrontierMath-Tiers-1-3-v2-Private
100.0·
FrontierMath-Tier-4-v2-Private
100.0·
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

99.7·
knowledgeCompare knowledge →
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

95.7·
SimpleQA Verified

Simple factual questions with verified correct answers. Tests accuracy of basic knowledge retrieval. Low scores indicate hallucination.

94.0·
Chess Puzzles

Tactical chess puzzles testing pattern recognition and multi-move calculation. Measures strategic reasoning ability.

77.9·
generalCompare general →
Ebr Bench
90.5·
Mystery Game Puzzles
89.0·
Furniture Assembly
85.7·
Quick compare:DeepSeek logovs DeepSeek V3.2 SpecialeOpenAI logovs GPT-6 Astrastepfun logovs Step 3.5 Flash
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
DeepSeek logo
DeepSeek V3.2 Speciale
DeepSeek
95.2$0.40/1M
OpenAI logo
GPT-6 Astra
OpenAI
93.4$10.00/1M
stepfun logo
Step 3.5 Flash
stepfun
89.5$0.10/1M
Links
Info
AnthropicPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Anthropic
BenchGecko API
claude-fable-5-2
Specifications
  • Typemultimodal
  • Context1.0M tokens (~500 books)
  • ReleasedNov 2026
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.056
Available On
Anthropic logoAnthropic$8.00
Categories
mathknowledgegeneral
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
DeepSeek logoDeepSeek V3.2 Speciale
95.2
OpenAI logoGPT-6 Astra
93.4
stepfun logoStep 3.5 Flash
89.5
OpenAI logoGPT-5 Chat
89.0
Claude Fable 5.2 is a proprietary multimodal AI model by Anthropic, released in November 2026. It has an average benchmark score of 95.4. Context window: 1M tokens.

Related Models

DeepSeek V3.2 Speciale · DeepSeekGPT-6 Astra · OpenAIStep 3.5 Flash · stepfunGPT-5 Chat · OpenAIClaude Opus 5.5 · Anthropic

Benchmarks

FrontierMath-Tiers-1-3-v2-PrivateFrontierMath-Tier-4-v2-PrivateOTIS Mock AIME 2024-2025GPQA diamondSimpleQA Verified

Related Pages

Anthropic · ProviderAnthropic · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-11-11

  • Claude Fable 5.2 by Anthropic. BenchGecko score 95.4, rank 1 of 326 scored models (normalized average of public benchmark scores).
  • List price $8.00 input · $40.00 output per 1M tokens (as of 2026-11-11).
  • Sold by 5 providers (as of 2026-11-11): Amazon Bedrock $8.00 in / $40.00 out · Anthropic $8.00 in / $40.00 out · Azure $8.00 in / $40.00 out · Google $8.00 in / $40.00 out · Anthropic $16.00 in / $80.00 out. Every provider
  • Gecko Tests: Who Are You B (Knows who made it) · Tokenizer Tax E (73% more tokens outside English).

How to cite · data as of 2026-11-11

Claude Fable 5.2 · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-11. https://benchgecko.ai/model/claude-fable-5-2

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP