Gemini 1.0 Pro Vision: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1527 ± 23, rank #1983 of 10916 rated models, from 154 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM VHELM - Mme - Celebrity | 92.65 | Prefix Quasi-Exact Match (%) | 96.8 |
| HELM VHELM - Blink Perception - Object Localization | 99.18 | Prefix Quasi-Exact Match (%) | 95.7 |
| HELM VHELM - Blink Perception - Relative Reflectance | 68.66 | Prefix Quasi-Exact Match (%) | 95.7 |
| HELM VHELM - Blink Perception - Visual Similarity | 97.78 | Prefix Quasi-Exact Match (%) | 92.6 |
| HELM VHELM - Blink Reasoning - Jigsaw | 98.67 | Prefix Quasi-Exact Match (%) | 89.4 |
| HELM VHELM - Exams V - French (Natural Science) | 84.21 | Prefix Quasi-Exact Match (%) | 86.2 |
| HELM Image2Struct - Image2Musicsheet | 40.24 | EMS | 84.2 |
| HELM VHELM - Blink Perception - Art Style | 94.87 | Prefix Quasi-Exact Match (%) | 83 |
| HELM VHELM - Blink Perception - Spatial Relation | 95.8 | Prefix Quasi-Exact Match (%) | 83 |
| HELM VHELM - Unicorn - Sketchy-VQA | 95.4 | Prefix Quasi-Exact Match (%) | 78.7 |
| HELM VHELM - Image2Webpage | 47.62 | EMD-Sim | 76.5 |
| HELM VHELM - Mmmu - Energy and Power | 50 | Prefix Quasi-Exact Match (%) | 75.5 |
Interactive version: theaggregate.ai/model?slug=gemini-1-0-pro-vision · How It Works · Data refreshed daily, snapshot 2026-11-08.