# MathVision → ScreenSpot Pro

Calibration · https://benchgap.net/calibration/mathvision/current/to/screenspot-pro/current · data retrieved 2026-11-21

ScreenSpot Pro is estimated from MathVision with a offset logistic curve fitted on 9 models measured on both: y = 0.5072 + (0.8822 − 0.5072) / (1 + exp(−31.73·(x − 0.8748))), R² = 0.88, cross-validated error 9.3 pp. It is used for 1 estimate.

| Estimated model | MathVision | ScreenSpot Pro | Source |
|---|---|---|---|
| [Gemma 4 12B](https://benchgap.net/model/gemma-4-12b.md) | 79.7% | 53.6% | estimated ± 9.3 pp, medium confidence |

## Fitted on

| Model | MathVision | ScreenSpot Pro |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 74.3% | 45.7% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 86.6% | 72.7% |
| [Inkling 2 Mini](https://benchgap.net/model/inkling-2-mini.md) | 89.0% | 76.5% |
| [Inkling 2 Small](https://benchgap.net/model/inkling-2-small.md) | 92.8% | 83.8% |
| [LFM3-VL-2B](https://benchgap.net/model/lfm3-vl-2b.md) | 51.6% | 55.5% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 88.6% | 65.6% |
| [Qwen3.6 Plus](https://benchgap.net/model/qwen3-6-plus.md) | 88.0% | 68.2% |
| [Qwen3.7 Plus](https://benchgap.net/model/qwen3-7-plus.md) | 90.3% | 79.0% |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 95.2% | 84.5% |
