Claude Fable 5.1 (xHigh): benchmark results
Provider: Anthropic. Released 2026-09-01. Access: API.
Aggregate Rating 1962 ± 15, rank #8 of 2168 rated models, from 61 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ErdosBench | 3.25 | Judge Average (0-4) | 100 |
| SEAL - SWE Atlas - Test Writing | 67.04 | Score | 100 |
| SciUniverse | 45.3 | Pass@1 (%) | 100 |
| SciUniverse - Assign structures from NMR | 29.2 | Pass@1 (%) | 100 |
| SciUniverse - Chemistry | 37.4 | Pass@1 (%) | 100 |
| SciUniverse - Develop a two-minute LC-MS method | 10 | Pass@1 (%) | 100 |
| SciUniverse - Manage PCR lab operations | 34.4 | Pass@1 (%) | 100 |
| SciUniverse - Materials | 68.8 | Pass@1 (%) | 100 |
| SciUniverse - Press BaTiO3 pellets | 100 | Pass@1 (%) | 100 |
| SciUniverse - Synthesize and detect an amide | 100 | Pass@1 (%) | 100 |
| SEAL - Humanity's Last Exam | 46.5 | Score | 98.1 |
| Conceptual Reasoning Index - Decision Theory (DTBench) | 96 | Chance-Corrected Score (0-100) | 97.9 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-1-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-20.