Claude Fable 5.1 (High): benchmark results
Provider: Anthropic. Released 2026-09-01. Access: API.
Unified ELO 1795 ± 1, rank #8 of 1761 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Computer Anthology Terminal Tasks (Claude Code) | 55.2 | pass@1 (%) | 100 |
| Computer Anthology Terminal Tasks (Terminus-2) | 62.4 | pass@1 (%) | 100 |
| WeirdML | 92.3 | Average Score | 99.4 |
| Creative Writing (Lechmazur) | 4.1 | Mean Score | 98.9 |
| Epoch AI - Critpt | 30.29 | Score | 98.2 |
| Agents on Rails | 92.1 | Successful Runs (%) | 96.9 |
| Epoch AI - Apex Agents | 44.4 | Score | 96.8 |
| Vals AI Terminal-Bench 2.1 | 85.02 | Accuracy (%) | 96.7 |
| ARC-AGI-2 | 88.75 | Accuracy (%) | 95.5 |
| Deep20Bench | 12.11 | Questions to Solve (mean, lower is better) | 93.8 |
| ARC-AGI-1 | 96 | Accuracy (%) | 92 |
| CursorBench 3.1 | 69.4 | Score (%) | 91.8 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-1-high · How It Works · Data refreshed daily, snapshot 2026-09-05.