Claude Opus 4.5 (20251101) (Thinking, High): benchmark results
Provider: Anthropic. Released 2025-11-24. Access: API.
Unified ELO 1762 ± 27, rank #209 of 2136 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Vals AI SAGE | 52.09 | Accuracy (%) | 92.2 |
| Vals AI Finance Agent | 58.81 | Accuracy (%) | 84.5 |
| Vals AI TaxEval v2 | 74.86 | Accuracy (%) | 81.9 |
| MedCode | 49.16 | Score (self-reported) | 81.1 |
| Vals AI LegalBench | 84.6 | Accuracy (%) | 81.1 |
| Vals AI MMLU-Pro | 87.26 | Accuracy (%) | 78.1 |
| Vals AI MedCode | 49.16 | Accuracy (%) | 78.1 |
| Vals AI MedScribe | 85.32 | Accuracy (%) | 74.8 |
| Vals AI CorpFin v2 | 65.07 | Accuracy (%) | 74.4 |
| ProofBench | 36 | Score (self-reported) | 70.7 |
| Vals AI GPQA | 85.86 | Accuracy (%) | 67.9 |
| Vals AI LiveCodeBench | 83.67 | Accuracy (%) | 67.6 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-5-20251101-thinking-high · How It Works · Data refreshed daily, snapshot 2026-10-10.