ScreenSpot-v2 (Mobile)
Multimodal · 2024-10-30
The mobile portion of ScreenSpot-v2: instructions on iOS and Android screenshots with text and icon targets. A prediction is correct when the click point falls inside the target element's bounding box.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemma 4 12B IT | 93.0 |
| Qwen3.5-4B | 92.4 |
| LFM3-VL-2B | 92.2 |
| GPT-5.6 Luna | 88.8 |
| Qwen3.5-2B | 82.2 |
| LFM2.5-VL-3B | 81.2 |
| Gemini 3.5 Flash-Lite | 80.0 |
| Gemma 4 E2B IT | 71.1 |