Claude Fable 5 (Low): benchmark results
Provider: Anthropic. Released 2026-06-09. Access: API.
Aggregate Rating 1832 ± 23, rank #98 of 2170 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CellTrack-Bench - Inspection F1 | 0.7 | Score (0–1) | 100 |
| CellTrack-Bench - Mean F1 | 0.76 | Score (0–1) | 100 |
| K-Bench - Risk (Therapeutic v0) | 95.17 | Risk Score (%, Therapeutic v0 prompt) | 100 |
| PointerBench - Average | 81.8 | Success rate (%) | 97.7 |
| PointerBench - Pro | 82.2 | Success rate (%) | 97.7 |
| PointerBench - Text | 67.4 | Success rate (%) | 97.7 |
| PointerBench - Sheets | 95.8 | Success rate (%) | 90.9 |
| CellTrack-Bench - Link F1 | 0.79 | Score (0–1) | 90.5 |
| CellTrack-Bench - Repair F1 | 0.79 | Score (0–1) | 90.5 |
| CellTrack-Bench - Subgraph LNK | 0.66 | Score (0–1) | 90.5 |
| K-Bench | 98.47 | Overall (%, Default v1 prompt) | 88.5 |
| OTIS Mock AIME 2024-25 | 97.78 | Accuracy (%) | 88.3 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-low · How It Works · Data refreshed daily, snapshot 2026-11-05.