Claude Opus 4.6 (Adaptive Reasoning, Max Effort): benchmark results
Claude Opus 4.6 evaluated in adaptive-reasoning mode at max effort. Provider: Anthropic. Released 2026-02-05. Access: API.
Aggregate Rating 1783 ± 19, rank #170 of 2168 rated models, from 30 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EnterpriseBench Corecraft (Paper Task Set) | 30.8 | Task pass rate (%) on the Corecraft benchmark task set evalu | 100 |
| UGI - Writing | 70.7 | Writing Score | 98.4 |
| UGI - Natural Intelligence | 70.78 | NatInt Score | 96.5 |
| Vals AI TaxEval v2 | 75.96 | Accuracy (%) | 94.8 |
| UGI Leaderboard | 52.2 | UGI Score | 92.9 |
| Vals AI CorpFin v2 | 67.02 | Accuracy (%) | 90.2 |
| Vals AI MMLU-Pro | 89.11 | Accuracy (%) | 89.8 |
| Inverse Turing Bench | 77.92 | Accuracy (imported) | 89.3 |
| Vals AI Finance Agent | 60.05 | Accuracy (%) | 89.3 |
| Vals AI SAGE | 51.58 | Accuracy (%) | 87.8 |
| Tau2-Bench Telecom | 92.11 | Success Rate (%) | 87.2 |
| Vals AI LegalBench | 85.3 | Accuracy (%) | 86 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-6-adaptive-reasoning-max-effort · How It Works · Data refreshed daily, snapshot 2026-10-20.