Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback): benchmark results
Provider: Anthropic. Released 2026-09-01. Access: API.
Unified ELO 1728 ± 1, rank #55 of 1761 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 84.67 | Accuracy (%) | 99.6 |
| AA Omniscience - Humanities & Social Sciences | 62.7 | Accuracy (%) | 99.2 |
| AA Omniscience - Science, Engineering & Mathematics | 58.2 | Accuracy (%) | 99.2 |
| AA-Omniscience Accuracy | 63.1 | Accuracy (%) | 99.2 |
| AA Omniscience - Business | 52.6 | Accuracy (%) | 99 |
| AA Omniscience - Law | 64.2 | Accuracy (%) | 99 |
| AA Humanity's Last Exam | 53.8 | Accuracy (%) | 98.7 |
| AA Omniscience - Health | 53.4 | Accuracy (%) | 98.6 |
| AA Omniscience | 37.6 | Score | 98.2 |
| AA Omniscience - Software Engineering (SWE) | 87.5 | Accuracy (%) | 98.2 |
| AA CritPt | 29.14 | Accuracy (%) | 98.1 |
| Artificial Analysis Intelligence Index | 49.87 | Intelligence Index | 97.5 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-1-adaptive-reasoning-medium-effort-default-fallback · How It Works · Data refreshed daily, snapshot 2026-09-05.