GUI Action Prediction on GUI-Odyssey
91.83Type Match (TM)OS-Atlas
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OS-AtlasModel Size=7B2025.06 | 91.83 | 76.76 | |
| OdysseyModel Size=7B2025.06 | 90.83 | 73.67 | |
| Mem-W-8BBackbone=UI-Venus-1.5-8B, Memory Augmentation=Mem-W2026.05 | 84.88 | 65.05 | |
| UI-Venus-1.5-8BModel Category=Baseline2026.05 | 83.12 | 62.45 | |
| Qwen3-VL-32BModel Category=Open-source2026.05 | 80.62 | 56.69 | |
| UI-Venus-1.5-30B-A3BModel Category=Open-source2026.05 | 80.56 | 60.64 | |
| GUI-Owl-1.5-8BModel Category=Open-source2026.05 | 79.24 | 58.95 | |
| Mobile-R1Model Size=3B2025.06 | 75.24 | 53.11 | |
| Mem-W-4BBackbone=Qwen3-VL-4B, Memory Augmentation=Mem-W2026.05 | 75.07 | 46.93 | |
| Qwen3-VL-8BModel Category=Open-source2026.05 | 74.96 | 48.73 | |
| Qwen3-VL-4BModel Category=Baseline2026.05 | 72.89 | 44.73 | |
| UI-TARS-1.5-7BModel Category=Open-source2026.05 | 72.13 | 50.98 | |
| MAI-UI-2BModel Category=Open-source2026.05 | 70.12 | 47.95 | |
| GLM-4.1V-9B-ThinkingModel Category=Open-source2026.05 | 67.26 | 32.88 | |
| Step-GUI-4BModel Category=Open-source2026.05 | 64.99 | 44.94 | |
| Qwen2.5-VLModel Size=7B2025.06 | 59.54 | 46.28 | |
| Qwen3-VL-2BModel Category=Open-source2026.05 | 58.85 | 38.7 | |
| AguvisModel Size=7B2025.06 | 26.71 | 13.54 | |
| Aguvis-7BModel Category=Open-source2026.05 | 26.7 | 13.5 | |
| OS-GenesisModel Size=7B2025.06 | 11.67 | 3.63 |