Claude Opus 4.7 (Non-reasoning): benchmark results
Provider: Anthropic. Released 2026-04-16. Access: API.
Aggregate Rating 1711 ± 24, rank #356 of 2286 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ALE-Bench | 1323.05 | Performance (Self-Refine x1) (imported) | 92.2 |
| WeirdML | 76.4 | Average Score | 85.1 |
| CoffeeBench | 2781.93 | Mean net income ($) | 83.3 |
| AstroAlertBench | 48.87 | End-to-end five-class accuracy (%; 1,500 ZTF alerts, 300 eac | 72.7 |
| BenchCAD 1.0 - Code Edit | 0.81 | Accuracy (0 to 1, share of the original-to-target IoU gap th | 66.7 |
| TUA-Bench | 49.7 | Success rate (%; mean of the verifier's scalar task reward o | 66.7 |
| DGEval - Regulatory Recall | 17.4 | Accuracy (%) | 61.8 |
| BenchCAD 1.0 - Code QA | 0.8 | Total accuracy (0 to 1) on 2,400 numeric CAD questions from | 55.6 |
| BenchCAD 1.0 - Vision QA | 0.53 | Total accuracy (0 to 1) on 2,400 numeric CAD questions from | 50 |
| DGEval - IMDG Code MCQ | 65.8 | Accuracy (%) | 38.2 |
| BenchCAD 1.0 - Vision2Code | 0.26 | IoU-score (voxel IoU x execution rate, 0 to 1), image to Cad | 36.4 |
| Generalization V2 (Lechmazur) | 52.6 | Inverse-Rank Score | 34.5 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-20.