Claude Opus 5 (Adaptive Reasoning, Low Effort): benchmark results
Provider: Anthropic. Released 2026-07-24. Access: API.
Unified ELO 1674 ± 1, rank #169 of 1761 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Science, Engineering & Mathematics | 54.01 | Accuracy (%) | 97.5 |
| AA Omniscience - Business | 48.4 | Accuracy (%) | 96.2 |
| AA Omniscience | 28.55 | Score | 95.9 |
| AA-Omniscience Accuracy | 55.95 | Accuracy (%) | 95.5 |
| AA Omniscience - Humanities & Social Sciences | 54.2 | Accuracy (%) | 95.4 |
| AA Humanity's Last Exam | 43.42 | Accuracy (%) | 94.8 |
| AA Omniscience - Health | 48.08 | Accuracy (%) | 94.5 |
| AA Omniscience - Software Engineering (SWE) | 79.2 | Accuracy (%) | 94.5 |
| AA CritPt | 23.14 | Accuracy (%) | 94.2 |
| AA Omniscience - Law | 52.3 | Accuracy (%) | 93.4 |
| AA Long Context Reasoning | 81.33 | Accuracy (%) | 93.3 |
| Artificial Analysis Intelligence Index | 43.81 | Intelligence Index | 93.2 |
Interactive version: theaggregate.ai/model?slug=claude-opus-5-adaptive-reasoning-low-effort · How It Works · Data refreshed daily, snapshot 2026-09-05.