Claude Fable 5 (Medium) — benchmark results

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

Unified ELO 2028 ± 45, rank #12 of 1844 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LisanBench0.89Mean Path Length / Current Maximum98
Epoch AI - ECI161.53ECI Score97.3
Sycophancy (Lechmazur)0.5Sycophancy rate % (lower is better)95.2
ARC-AGI-282.5Accuracy (%)91.8
ARC-AGI-192.5Accuracy (%)85.8
CursorBench 3.165.2Score (%)82.9
Vending-Bench 24339.81Money Balance ($)52.6

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