Claude Fable 5.1 (Low): benchmark results
Provider: Anthropic. Released 2026-09-01. Access: API.
Aggregate Rating 1878 ± 23, rank #59 of 2170 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DuelLab GameBench 2 - Reasoning Variants | 84.8 | GameBench2 normalized score (0–100) | 94.6 |
| Conceptual Reasoning Index - Argument Evaluation (LMCA) | 63.46 | Chance-Corrected Score (0-100) | 91.9 |
| Land or Water? (Celeste) | 98.2 | Area-weighted accuracy (%) | 90.9 |
| Epoch AI - Critpt | 27.71 | Score | 88.2 |
| Conceptual Reasoning Index | 72.27 | Chance-Corrected Score (0-100) | 87.6 |
| Conceptual Reasoning Index - Consistency (ACCoRD) | 80.76 | Chance-Corrected Score (0-100) | 86.7 |
| ARC-AGI-2 | 78.33 | Accuracy (%) | 85.9 |
| MathArena - ARXIV_FALSE June | 86.57 | Accuracy (%) | 79.3 |
| Conceptual Reasoning Index - Decision Theory (DTBench) | 90.22 | Chance-Corrected Score (0-100) | 78.5 |
| MathArena Arxiv False | 78 | Accuracy (%) | 77.3 |
| MathArena - ARXIV_FALSE August | 77.98 | Accuracy (%) | 76.5 |
| Epoch AI - Dtbench | 94.13 | Score | 75.4 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-1-low · How It Works · Data refreshed daily, snapshot 2026-11-05.