Claude Opus 4.7 (Low): benchmark results
Provider: Anthropic. Released 2026-04-16. Access: API.
Aggregate Rating 1689 ± 23, rank #419 of 2286 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BAGEN - SWE-bench - Interval Reward | 0.16 | Mean interval score (coverage x tightness, 0-1) | 100 |
| BAGEN - Search-R1 - F1@All | 40.5 | All-turn feasibility macro-F1 (%) | 100 |
| TeachObs - Scene Coding (Transcript) | 27.7 | Macro F1 (%; 39 observation codes per 15-second classroom sc | 100 |
| BAGEN - SWE-bench - F1@All | 51.1 | All-turn feasibility macro-F1 (%) | 75 |
| BAGEN - Search-R1 - Interval Reward | 0.11 | Mean interval score (coverage x tightness, 0-1) | 75 |
| BAGEN - Warehouse - Interval Reward | 0.69 | Mean interval score (coverage x tightness, 0-1) | 75 |
| TeachObs - Lesson Narrative Coverage | 29.5 | Reference coverage (%; share of the experts' atomic claims a | 75 |
| TeachObs - Scene Coding (Transcript + Frame) | 43.8 | Macro F1 (%; 39 observation codes per 15-second classroom sc | 75 |
| DGEval - IMDG Code MCQ | 68.3 | Accuracy (%) | 58.8 |
| DGEval - Regulatory Recall | 16.2 | Accuracy (%) | 55.9 |
| BAGEN - Sokoban - F1@All | 45.6 | All-turn feasibility macro-F1 (%) | 50 |
| Equation-Suffix Prediction (Kimi K2.6 Scorer) | 0.15 | Likelihood lift (mean clipLL2 per target token over the same | 50 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7-low · How It Works · Data refreshed daily, snapshot 2026-11-20.