GUI Localization on GUIZoom-Bench 1.0 (hard-normal)
69.6Localization AccuracyQwen3-VL-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-VL-32BZoom Depth=d22025.12 | 69.6 | |
| Qwen3-VL-32BZoom Depth=d32025.12 | 67.3 | |
| Qwen3-VL-32BZoom Depth=d42025.12 | 67.3 | |
| UI-Venus-7BZoom Depth=d42025.12 | 57.6 | |
| Qwen3-VL-8BZoom Depth=d42025.12 | 56.7 | |
| Qwen3-VL-8BZoom Depth=d22025.12 | 55.3 | |
| Qwen3-VL-8BZoom Depth=d32025.12 | 55.3 | |
| UI-Venus-7BZoom Depth=d32025.12 | 54.8 | |
| UI-Venus-7BZoom Depth=d22025.12 | 47 | |
| UI-TARS-1.5-7BZoom Depth=d42025.12 | 39.2 | |
| UI-TARS-1.5-7BZoom Depth=d32025.12 | 37.3 | |
| UI-TARS-1.5-7BZoom Depth=d22025.12 | 35 | |
| UI-TARS-7BZoom Depth=d42025.12 | 32.3 | |
| Qwen3-VL-32BZoom Depth=d12025.12 | 30.4 | |
| UI-TARS-7BZoom Depth=d22025.12 | 30.4 | |
| UI-TARS-7BZoom Depth=d32025.12 | 30.4 | |
| Qwen2.5-VL-7BZoom Depth=d42025.12 | 28.6 | |
| Qwen2.5-VL-7BZoom Depth=d32025.12 | 28.1 | |
| UI-Venus-7BZoom Depth=d12025.12 | 28.1 | |
| Qwen3-VL-8BZoom Depth=d12025.12 | 24.9 | |
| UI-TARS-1.5-7BZoom Depth=d12025.12 | 21.8 | |
| Qwen2.5-VL-7BZoom Depth=d22025.12 | 21.7 | |
| Qwen2.5-VL-7BZoom Depth=d12025.12 | 12.9 | |
| UI-TARS-7BZoom Depth=d12025.12 | 11.5 | |
| GPT-5Zoom Depth=d42025.12 | 6.9 | |
| GPT-5Zoom Depth=d32025.12 | 6.5 | |
| GPT-5Zoom Depth=d12025.12 | 5.5 | |
| GPT-5Zoom Depth=d22025.12 | 5.5 | |
| Claude-4.5-SonnetZoom Depth=d12025.12 | 4.9 |