Claude Opus 4.7 (Adaptive Reasoning, Max Effort) — benchmark results
Claude Opus 4.7 evaluated in adaptive-reasoning mode at max effort. Provider: Anthropic. Released 2026-04-16. Access: API.
Unified ELO 1936 ± 35, rank #45 of 1776 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience | 26.17 | Score | 99.1 |
| AA Omniscience - Software Engineering (SWE) - Julia | 76 | Accuracy (%) | 98.3 |
| Artificial Analysis Intelligence Index | 53.53 | Intelligence Index | 98.2 |
| AA Omniscience - Software Engineering (SWE) - Swift | 84 | Accuracy (%) | 98.1 |
| AA SciCode | 54.51 | Accuracy (%) | 97.4 |
| AA Omniscience - Software Engineering (SWE) - Go | 78 | Accuracy (%) | 97.3 |
| AA Omniscience - Software Engineering (SWE) - Java | 62 | Accuracy (%) | 96.9 |
| AA Omniscience - Software Engineering (SWE) - Dart | 66 | Accuracy (%) | 96.5 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 70 | Accuracy (%) | 96.4 |
| AA GPQA Diamond | 91.41 | Accuracy (%) | 96.3 |
| AA Humanity's Last Exam | 39.62 | Accuracy (%) | 96.1 |
| AA Omniscience - Software Engineering (SWE) | 72 | Accuracy (%) | 95.8 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7-adaptive-reasoning-max-effort · How the rankings work · Data refreshed daily, snapshot 2026-07-22.