Claude Fable 5 (Max): benchmark results

Claude Fable 5 evaluated at the max reasoning-effort setting. Provider: Anthropic. Released 2026-06-09. Access: API.

Unified ELO 1775 ± 1, rank #16 of 1761 rated models, from 86 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Epoch AI - Dtbench98.4Score100
LiveBench Code Generation91.55Score100
MathArena - ARXIV April70.73Accuracy (%)100
MathArena Arxiv87.5Accuracy (%)100
PostTrainBench41.79Weighted Avg Score100
Vals AI LegalBench88.56Accuracy (%)100
Vals AI Multimodal Index74.15Accuracy (%)100
Vals AI Vibe Code Bench90.35Accuracy (%)100
Vals AI Web Search48.45Accuracy (%)100
Vals Index75.14Score (self-reported)100
Vals Multimodal Index74.15Score (self-reported)100
Conceptual Reasoning Index - Argument Evaluation (LMCA)60.34Chance-Corrected Score (0-100)99.4

Interactive version: theaggregate.ai/model?slug=claude-fable-5-max · How It Works · Data refreshed daily, snapshot 2026-09-05.