Claude Opus 4.6 (Max): benchmark results
Claude Opus 4.6 evaluated at the max reasoning-effort setting. Provider: Anthropic. Released 2026-02-05. Access: API.
Aggregate Rating 1782 ± 9, rank #171 of 2168 rated models, from 162 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Global-MMLU-Lite - Chinese | 93.58 | Accuracy (%) | 100 |
| AA Global-MMLU-Lite - English | 95.17 | Accuracy (%) | 100 |
| AA Global-MMLU-Lite - Korean | 93 | Accuracy (%) | 100 |
| ATLAS | 70.55 | ATLAScore (self-reported) | 100 |
| BC-VL | 51.5 | Score (%) | 100 |
| Global PIQA | 91.2 | Score (%) | 100 |
| KellyBench - Strategy Sophistication | 26.5 | Strategy sophistication (% of the 52-point expert rubric of | 100 |
| Kernel Bench L3 | 2.63 | Median speedup (x) | 100 |
| NOVA-63 | 59.1 | Score (%) | 100 |
| Odysseys | 44.5 | Perfect-rubric rate (%: share of tasks with every rubric ite | 100 |
| Odysseys - Easy | 97.8 | Perfect-rubric rate (%: share of tasks with every rubric ite | 100 |
| Odysseys - Hard | 11 | Perfect-rubric rate (%: share of tasks with every rubric ite | 100 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-6-max · How It Works · Data refreshed daily, snapshot 2026-10-20.