Claude Fable 5.2 (High): benchmark results
Provider: Anthropic. Released 2026-11-09. Access: API.
Aggregate Rating 2014 ± 18, rank #3 of 2201 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HalachaBench | 85.77 | Mean rubric score on 60 semicha exam questions, three attemp | 100 |
| The Aggregate - Exotic Fact Recall | 75 | Accuracy (%) | 100 |
| The Aggregate - Long-tail Book Recall | 77 | Accuracy (%) | 100 |
| Vals AI (Vals Index) | 71.31 | Accuracy (%) | 100 |
| Vals AI BioMysteryBench | 83.7 | Accuracy (%) | 100 |
| Vals AI Excel Modeling v1.1 | 77.38 | Accuracy (%) | 100 |
| Vals AI Finance Agent v2.1 | 69.47 | Dealbreaker-gated Partial Credit (%) | 100 |
| Vals AI ProgramBench | 87.79 | Raw Pass Rate (%) | 100 |
| Vals AI Vibe Code Bench | 93.77 | Accuracy (%) | 100 |
| Vals AI Terminal-Bench 4.0 | 65.15 | Accuracy (%) | 99.2 |
| Conceptual Reasoning Index | 78.92 | Chance-Corrected Score (0-100) | 99.1 |
| Conceptual Reasoning Index - Consistency (ACCoRD) | 88.03 | Chance-Corrected Score (0-100) | 99.1 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-2-high · How It Works · Data refreshed daily, snapshot 2026-11-12.