GUI Interaction Navigation on CMGUI
96.4Step AccuracySecAgent-3B
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SecAgent-3BModel Category=Open-source Models (3B and 4B)2026.03 | 96.4 | 80 | 96.1 | 99.1 | 87.6 | 99.2 | |
| Qwen3-VL-8BModel Category=Open-source Models (7B and 8B)2026.03 | 91.1 | 59.7 | 92.6 | 82.6 | 69.2 | 94.6 | |
| Qwen3-VL-4BModel Category=Open-source Models (3B and 4B)2026.03 | 90.1 | 57.4 | 91.2 | 84.8 | 65.9 | 94.6 | |
| UI-Venus-Navi-7BModel Category=Open-source Models (7B and 8B)2026.03 | 89.6 | 53.1 | 92.6 | 91.5 | 45 | 81.2 | |
| AgentCPM-GUI-8BModel Category=Open-source Models (7B and 8B)2026.03 | 84.9 | 39 | 85.8 | 93.8 | 54.9 | 81.2 | |
| Qwen2.5-VL-7BModel Category=Open-source Models (7B and 8B)2026.03 | 83.8 | 35.6 | 89 | 75.9 | 4.4 | 79 | |
| Claude-3-SonnetModel Category=Closed-source Models2026.03 | 81.4 | 25.8 | 91.1 | 30 | 81.3 | 53.4 | |
| Qwen2.5-VL-3BModel Category=Open-source Models (3B and 4B)2026.03 | 78.8 | 25.6 | 83.2 | 66 | 5.5 | 81.9 | |
| GUI-R1-7BModel Category=Open-source Models (7B and 8B)2026.03 | 70.5 | 7.9 | 84.2 | 45.5 | 5.5 | 8.7 | |
| OS-Atlas-7BModel Category=Open-source Models (7B and 8B)2026.03 | 69.6 | 10 | 77 | 76.8 | 5.5 | 32.1 | |
| UI-TARS-7BModel Category=Open-source Models (7B and 8B)2026.03 | 63.8 | 2.5 | 80.2 | 11.1 | 3.3 | 8.3 | |
| UI-R1-3BModel Category=Open-source Models (3B and 4B)2026.03 | 56.4 | 1 | 71.2 | 7.6 | 5.5 | 6.5 | |
| GPT-4oModel Category=Closed-source Models2026.03 | 48.4 | 0 | 43.5 | 68.3 | 63.7 | 62.1 | |
| GUI-R1-3BModel Category=Open-source Models (3B and 4B)2026.03 | 45.2 | 0.5 | 57.1 | 8 | 2.2 | 4.3 |