Claude Fable 5 (Low) — benchmark results

Provider: Anthropic. Released 2026-06-09. Access: API.

Unified ELO 2007 ± 34, rank #13 of 1844 rated models, from 8 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Epoch AI - ECI161.53ECI Score97.3
InferenceBench8.74Speedup Score96.4
ARC-AGI-276.81Accuracy (%)90.7
Opus Magnum Bench49.28Human-normalized score (%)85.7
ARC-AGI-190.5Accuracy (%)79.7
CursorBench 3.162.1Score (%)68.3
Vending-Bench 25018.52Money Balance ($)61.5
Epoch AI - Mystery Game Puzzles15Score46.3

Interactive version: theaggregate.ai/model?slug=claude-fable-5-low · How It Works · Data refreshed daily, snapshot 2026-08-06.