AutoGUI-v2 - Region Grounding (Functionality): leaderboard
Metric: Region grounding by functionality on 442 tasks, grounding center accuracy (%): the predicted point falls inside the ground-truth box; native resolution, the authors' tuned prompts; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-4.5V | 84.6 |
| 2 | Qwen 3 VL 32B Instruct | 84.4 |
| 3 | Gemini 2.5 Pro | 77.6 |
| 4 | uitars-1.5-7B | 75.8 |
| 5 | Qwen 3 VL 8B Instruct | 70.1 |
| 6 | Claude Sonnet 4.5 | 48.6 |
| 7 | O3 | 36.7 |
| 8 | Step-3 | 36.7 |
| 9 | GPT-5 | 31.4 |
Interactive version: theaggregate.ai/benchmark?slug=autogui-v2-region-grounding-functionality · How It Works · Data refreshed daily, snapshot 2026-10-07.