Claude Opus 5 (Adaptive Reasoning, Medium Effort): benchmark results
Provider: Anthropic. Released 2026-07-24. Access: API.
Unified ELO 1701 ± 1, rank #105 of 1761 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Humanity's Last Exam | 51.3 | Accuracy (%) | 98 |
| AA Omniscience | 31.02 | Score | 97.1 |
| AA Omniscience - Science, Engineering & Mathematics | 53.91 | Accuracy (%) | 97.1 |
| Artificial Analysis Intelligence Index | 49.52 | Intelligence Index | 97.1 |
| AA Omniscience - Business | 49.4 | Accuracy (%) | 96.7 |
| AA Omniscience - Health | 51.16 | Accuracy (%) | 96.7 |
| AA-Omniscience Accuracy | 57.07 | Accuracy (%) | 96.1 |
| AA CritPt | 26.86 | Accuracy (%) | 95.9 |
| AA Omniscience - Humanities & Social Sciences | 54.2 | Accuracy (%) | 95.4 |
| AA Long Context Reasoning | 82 | Accuracy (%) | 94.9 |
| AA Omniscience - Software Engineering (SWE) | 81.5 | Accuracy (%) | 94.9 |
| AA Omniscience - Law | 52.7 | Accuracy (%) | 93.6 |
Interactive version: theaggregate.ai/model?slug=claude-opus-5-adaptive-reasoning-medium-effort · How It Works · Data refreshed daily, snapshot 2026-09-05.