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

BenchmarkScoreMetricPercentile
AA Omniscience - Science, Engineering & Mathematics54.01Accuracy (%)97.5
AA Omniscience - Business48.4Accuracy (%)96.2
AA Omniscience28.55Score95.9
AA-Omniscience Accuracy55.95Accuracy (%)95.5
AA Omniscience - Humanities & Social Sciences54.2Accuracy (%)95.4
AA Humanity's Last Exam43.42Accuracy (%)94.8
AA Omniscience - Health48.08Accuracy (%)94.5
AA Omniscience - Software Engineering (SWE)79.2Accuracy (%)94.5
AA CritPt23.14Accuracy (%)94.2
AA Omniscience - Law52.3Accuracy (%)93.4
AA Long Context Reasoning81.33Accuracy (%)93.3
Artificial Analysis Intelligence Index43.81Intelligence Index93.2

Interactive version: theaggregate.ai/model?slug=claude-opus-5-adaptive-reasoning-low-effort · How It Works · Data refreshed daily, snapshot 2026-09-05.