Claude Sonnet 5 (Adaptive Reasoning, Medium Effort): benchmark results
Provider: Anthropic. Released 2026-06-30. Access: API.
Unified ELO 1613 ± 1, rank #402 of 1919 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Writing | 52.4 | Writing Score | 89.1 |
| UGI - Natural Intelligence | 40.16 | NatInt Score | 86.4 |
| Artificial Analysis Intelligence Index | 28.44 | Intelligence Index | 85.7 |
| AA CritPt | 8.57 | Accuracy (%) | 83.3 |
| AA Omniscience - Software Engineering (SWE) | 57 | Accuracy (%) | 82.6 |
| AA Humanity's Last Exam | 29.98 | Accuracy (%) | 81.9 |
| AA Omniscience - Science, Engineering & Mathematics | 41.6 | Accuracy (%) | 80.5 |
| AA Omniscience - Business | 29.2 | Accuracy (%) | 79.7 |
| AA Omniscience - Humanities & Social Sciences | 35 | Accuracy (%) | 79.6 |
| AA-Omniscience Accuracy | 37.1 | Accuracy (%) | 79.3 |
| AA Omniscience - Health | 33.1 | Accuracy (%) | 77.8 |
| AA Omniscience - Law | 26.7 | Accuracy (%) | 77.6 |
Interactive version: theaggregate.ai/model?slug=claude-sonnet-5-adaptive-reasoning-medium-effort · How It Works · Data refreshed daily, snapshot 2026-09-08.