Claude Fable 5 (Medium): benchmark results
Provider: Anthropic. Released 2026-06-09. Access: API.
Unified ELO 1736 ± 1, rank #49 of 3078 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LisanBench | 0.89 | Mean Path Length / Current Maximum | 98 |
| Conceptual Reasoning Index - Consistency (ACCoRD) | 82.37 | Chance-Corrected Score (0-100) | 97.8 |
| Conceptual Reasoning Index - Argument Evaluation (LMCA) | 60.32 | Chance-Corrected Score (0-100) | 91.9 |
| Conceptual Reasoning Index - Decision Theory (DTBench) | 92.88 | Chance-Corrected Score (0-100) | 90.7 |
| Conceptual Reasoning Index | 71.24 | Chance-Corrected Score (0-100) | 90.2 |
| ARC-AGI-2 | 82.5 | Accuracy (%) | 88.3 |
| DataBench | 76 | Score (%) | 87.1 |
| ARC-AGI-1 | 92.5 | Accuracy (%) | 83.9 |
| DuelLab Overall | 49.6 | DuelLab Score | 82.2 |
| ObviousBench | 97.92 | Answer pass³ (%) | 80.9 |
| Maze-Bench | 39.35 | Mean Score | 78.1 |
| Epoch AI - Cursorbench | 65.2 | Score | 76.7 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-medium · How It Works · Data refreshed daily, snapshot 2026-09-20.