Keye-VL-8B-Preview: benchmark results
Provider: Other.
Aggregate Rating 1532 ± 49, rank #1926 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| E-VAds (Relaxed-5) - Regulatory Compliance | 70.7 | Relaxed-5 score (%) on the Regulatory Compliance task (poten | 100 |
| E-VAds (Relaxed-5) - Cross-Modal Detection | 71.2 | Relaxed-5 score (%) on the Cross-Modal Detection task (consi | 87.5 |
| VRBench | 60.44 | Overall (imported) | 75 |
| E-VAds (Relaxed-5) | 57.2 | Relaxed-5 score (%): mean judge score over all five tasks; 1 | 25 |
| E-VAds | 11.9 | Strict score (%): share of answers the judge rates fully cor | 12.5 |
| E-VAds (Relaxed-5) - Consumer Insight | 46.2 | Relaxed-5 score (%) on the Consumer Insight task (the target | 12.5 |
| MME-CC | 9.65 | Overall Score (%) | 6.7 |
| E-VAds (Relaxed-5) - Basic Perception | 56.8 | Relaxed-5 score (%) on the Basic Perception task (product at | 0 |
| E-VAds (Relaxed-5) - Marketing Logic | 38.6 | Relaxed-5 score (%) on the Marketing Logic task (selling poi | 0 |
Interactive version: theaggregate.ai/model?slug=keye-vl-8b-preview · How It Works · Data refreshed daily, snapshot 2026-11-08.