tensor.news
Integrity IndexModelsCompareLabsPapersMethodologyAbout
tensor.news

© 2026

Evidence

  • Integrity Index
  • Model Claim Ledger
  • Compare

Entities

  • Models
  • Benchmarks
  • Labs
  • Papers

Method

  • Methodology
  • About
  • Privacy

Machine

  • Feed (Atom)
  • llms.txt
  • llms-full.txt
  • Sitemap

tensor.news grades benchmarks and tags model claims by evidence status. Every fact carries a source record and provenance. Scores are framed as task performance under a disclosed harness, never as deployed capability.

← Models

Claude Fable 5

Anthropic·released 2026-06-09

independently reproduced

Scored on 12 benchmarks, leading 6. 42% of its scores are independently reproduced — the rest are self-reported or unverified.

12
Benchmarks scored
6
SOTA leads
99.72
Best score
42%
Reproduced
5 of 12

Leads on: CursorBench, FrontierMath-Tier-4-v2-Private, GBAEval, Remote Labor Index, SimpleBench, WeirdML

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • OTIS Mock AIME 2024-2025
    99.72
  • WeirdML
    87.85
  • FrontierMath-Tier-4-v2-Private
    87.8
  • FrontierMath-Tiers-1-3-v2-Private
    87.02
  • SimpleBench
    78.28
  • GBAEval
    74.47
  • CursorBench
    72.9
  • SimpleQA Verified
    68.3
  • APEX-Agents
    45
  • Chess Puzzles
    41
  • CritPt
    28.57
  • Remote Labor Index
    16.1

All benchmark scores

BenchmarkScoreEvidence statusMeasured
OTIS Mock AIME 2024-202599.72accuracy (%)reproduced· optimized2026-06-09
WeirdML87.85accuracy (%)unverified2026-06-09
FrontierMath-Tier-4-v2-Private87.8accuracy (%)reproduced2026-06-09
FrontierMath-Tiers-1-3-v2-Private87.02accuracy (%)reproduced2026-06-09
SimpleBench78.28accuracy (%)unverified2026-06-09
GBAEval74.47accuracy (%)unverified2026-06-09
CursorBench72.9accuracy (%)unverified2026-06-09
SimpleQA Verified68.3accuracy (%)reproduced2026-06-09
APEX-Agents45accuracy (%)unverified2026-06-09
Chess Puzzles41accuracy (%)reproduced2026-06-09
CritPt28.57accuracy (%)unverified2026-06-09
Remote Labor Index16.1accuracy (%)unverified2026-06-09
12 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from Anthropic

  • Claude Opus 4.898.33
  • Claude Opus 4.797.8
  • Claude Sonnet 4.597.73
  • Claude Haiku 4.596.36
  • Claude Opus 4.694.44
  • Claude 3.7 Sonnet91.16

Head-to-head

  • vs PowerMoE-3b
  • vs GPT-5.5
  • vs Gemini 3.1 Pro
  • vs Llama 3.1-405B
  • vs phi-3-small 7.4B
  • vs DeepSeek-V3

Benchmarks it leads

  • CursorBench
  • FrontierMath-Tier-4-v2-Private
  • GBAEval
  • Remote Labor Index
  • SimpleBench
  • WeirdML