GUI Grounding on OSWorld-G v1.0 (test)
67.6AccuracyVISTA
Evaluation Results
| Method | Links | |
|---|---|---|
| VISTAModel Scale=≥30B, Inference Method=Multi-view (+ MVP)2026.06 | 67.6 | |
| VISTAModel Scale=≥30B, Inference Method=Single-view2026.06 | 67.1 | |
| Qwen3-VLModel Scale=≥30B2026.06 | 66.5 | |
| Holo2Model Scale=≥30B2026.06 | 65.2 | |
| GTA1Model Scale=≥30B2026.06 | 65.2 | |
| VISTAModel Scale=≈4B, Inference Method=Multi-view (+ MVP)2026.06 | 64 | |
| VISTAModel Scale=≈4B, Inference Method=Single-view2026.06 | 63.8 | |
| Holo2Model Scale=≈8B2026.06 | 63.5 | |
| VISTAModel Scale=≈8B, Inference Method=Multi-view (+ MVP)2026.06 | 63.1 | |
| VISTAModel Scale=≈8B, Inference Method=Single-view2026.06 | 62.4 | |
| UI-VenusModel Scale=72B2026.06 | 62.2 | |
| Step-GUIModel Scale=≈4B2026.06 | 60.5 | |
| GTA1Model Scale=≈8B2026.06 | 60.1 | |
| MAI-UIModel Scale=≈8B2026.06 | 60.1 | |
| OpenCUAModel Scale=≥30B2026.06 | 59.6 | |
| Qwen3-VLModel Scale=≈4B2026.06 | 58.2 | |
| GUI-OwlModel Scale=≥30B2026.06 | 58 | |
| GUI-OwlModel Scale=≈8B2026.06 | 55.9 | |
| OpenCUAModel Scale=≈8B2026.06 | 55.3 | |
| Qwen3-VLModel Scale=≈8B2026.06 | 54.8 | |
| UI-VenusModel Scale=≈8B2026.06 | 54.6 | |
| UI-TARS-1.5Model Scale=≈8B2026.06 | 52.8 |