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
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Epoch AI - Dtbench | 98.4 | Score | 100 |
| LiveBench Code Generation | 91.55 | Score | 100 |
| MathArena - ARXIV April | 70.73 | Accuracy (%) | 100 |
| MathArena Arxiv | 87.5 | Accuracy (%) | 100 |
| PostTrainBench | 41.79 | Weighted Avg Score | 100 |
| Vals AI LegalBench | 88.56 | Accuracy (%) | 100 |
| Vals AI Multimodal Index | 74.15 | Accuracy (%) | 100 |
| Vals AI Vibe Code Bench | 90.35 | Accuracy (%) | 100 |
| Vals AI Web Search | 48.45 | Accuracy (%) | 100 |
| Vals Index | 75.14 | Score (self-reported) | 100 |
| Vals Multimodal Index | 74.15 | Score (self-reported) | 100 |
| Conceptual Reasoning Index - Argument Evaluation (LMCA) | 60.34 | Chance-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.