GUI Grounding on ScreenSpot Pro (Category Metrics)
7,350Average ScoreMAI-UI-32B
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MAI-UI-32BZoom-In=true2025.12 | 7,350 | — | — | — | — | — | — | 84.4 | 57.9 | 87.9 | 46.2 | 79.2 | 53.1 | 91.7 | 54.5 | 88.1 | 79.2 | 80.4 | 47.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Seed1.82026.02 | 7,310 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Seed1.8Type=Proprietary2025.12 | 7,310 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Gemini-3-Pro2026.02 | 7,270 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Gemini-3-ProType=Proprietary2025.12 | 7,270 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-8BZoom-In=true2025.12 | 7,090 | — | — | — | — | — | — | 78.6 | 58.6 | 78.8 | 46.9 | 80.7 | 43.8 | 86.1 | 49.1 | 88.1 | 81.1 | 76.6 | 51.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Triple-Step CoGConfiguration=UI-TARS1.5-7B -> Qwen3-VL-235B -> Qwen3-VL-32B, own evaluation=true2025.12 | 6,840 | 6,920 | 6,570 | 57.9 | 69.3 | 8,090 | 7,190 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-32BZoom-In=false2025.12 | 6,790 | — | — | — | — | — | — | 86.4 | 40.7 | 82.8 | 37.8 | 70.1 | 45.3 | 91.7 | 46.4 | 90.4 | 71.7 | 78.5 | 34.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Dual-Step CoGConfiguration=two instances of Qwen3-VL-235B, own evaluation=true2025.12 | 6,670 | 6,640 | 6,310 | 59.5 | 70.9 | 8,090 | 6,230 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-8BModel size=8B2026.02 | 6,580 | — | — | — | — | — | — | 83.8 | 52.4 | 76.3 | 33.6 | 72.6 | 35.9 | 79.9 | 37.3 | 88.7 | 60.4 | 76.6 | 49.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-8BZoom-In=false2025.12 | 6,580 | — | — | — | — | — | — | 83.8 | 52.4 | 76.3 | 33.6 | 72.6 | 35.9 | 79.9 | 37.3 | 88.7 | 60.4 | 76.6 | 49.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-235B-InstructModel Type=Open-Source, own evaluation=true2025.12 | 6,390 | 6,640 | 6,310 | 52.8 | 66.1 | 7,580 | 6,190 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTA-32BModel Type=Open-Source2025.12 | 6,360 | 6,120 | 5,280 | 60.5 | 65 | 8,350 | 6,530 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTA1-32B2025.12 | 6,360 | — | — | — | — | — | — | 82.5 | 28.3 | 69.2 | 14.7 | 43.7 | 23.4 | 79.9 | 31.8 | 80.8 | 43.4 | 70.1 | 32.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Holo1.5-72BModel Type=Open-Source2025.12 | 6,330 | 6,350 | 6,250 | 51.3 | 64.2 | 7,960 | 5,970 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-2BZoom-In=true2025.12 | 6,280 | — | — | — | — | — | — | 75.3 | 42.8 | 74.7 | 30.1 | 69.5 | 34.4 | 84 | 42.7 | 85.3 | 56.6 | 69.2 | 47.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-Venus-72B2025.12 | 6,190 | — | — | — | — | — | — | 84.4 | 33.1 | 73.2 | 30.8 | 66.5 | 29.7 | 84.7 | 42.7 | 83.1 | 60.4 | 75.7 | 36 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-1.5Model Type=Proprietary2025.12 | 6,160 | 6,390 | 5,040 | 58.2 | 69.3 | 7,960 | 5,100 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-32B-InstructModel Type=Open-Source, own evaluation=true2025.12 | 6,140 | 6,750 | 5,760 | 47.9 | 65.3 | 7,360 | 6,010 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Seed-1.5-VLModel Type=Proprietary2025.12 | 6,090 | 5,380 | 5,920 | 59 | 61.4 | 7,480 | 6,020 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| POINTS-GUI-G-8BModel size=8B2026.02 | 5,990 | — | — | — | — | — | — | 76 | 44.8 | 66.2 | 33.6 | 46.7 | 28.1 | 81.3 | 48.2 | 88.7 | 62.3 | 72 | 43.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Owl-32BModel size=32B2026.02 | 5,800 | — | — | — | — | — | — | 84.4 | 39.3 | 65.2 | 18.2 | 62.4 | 28.1 | 82.6 | 39.1 | 81.4 | 39.6 | 70.1 | 36 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Owl-32B2025.12 | 5,800 | — | — | — | — | — | — | 84.4 | 39.3 | 65.2 | 18.2 | 62.4 | 28.1 | 82.6 | 39.1 | 81.4 | 39.6 | 70.1 | 36 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-2BModel size=2B2026.02 | 5,740 | — | — | — | — | — | — | 76.6 | 32.4 | 69.2 | 21.7 | 61.4 | 23.4 | 81.2 | 34.5 | 85.9 | 39.6 | 68.2 | 41.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAI-UI-2BZoom-In=false2025.12 | 5,740 | — | — | — | — | — | — | 76.6 | 32.4 | 69.2 | 21.7 | 61.4 | 23.4 | 81.2 | 34.5 | 85.9 | 39.6 | 68.2 | 41.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenCUA-32BModel size=32B2026.02 | 5,530 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenCUA-32B2025.12 | 5,530 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Owl-7BModel size=7B2026.02 | 5,490 | — | — | — | — | — | — | 76.6 | 31 | 59.6 | 27.3 | 64.5 | 21.9 | 79.1 | 37.3 | 77.4 | 39.6 | 59.8 | 33.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-32BModel size=32B2026.02 | 5,490 | — | — | — | — | — | — | 69.5 | 22.1 | 75.8 | 25.2 | 60.4 | 28.1 | 84.7 | 25.5 | 85.9 | 43.4 | 62.6 | 15.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Owl-7B2025.12 | 5,490 | — | — | — | — | — | — | 76.6 | 31 | 59.6 | 27.3 | 64.5 | 21.9 | 79.1 | 37.3 | 77.4 | 39.6 | 59.8 | 33.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-32BEvaluated by us=true2025.12 | 5,490 | — | — | — | — | — | — | 69.5 | 22.1 | 75.8 | 25.2 | 60.4 | 28.1 | 84.7 | 25.5 | 85.9 | 43.4 | 62.6 | 15.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-72B-InstructModel Type=Open-Source2025.12 | 5,330 | 5,350 | 4,490 | 44.4 | 59.1 | 7,260 | 4,950 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-UI Lite2026.02 | 5,330 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-UI Lite2025.12 | 5,330 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-SpotlightModel Type=Open-Source2025.12 | 5,280 | 5,260 | 4,480 | 51.6 | 53.5 | 7,040 | 4,590 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InfiGUI-G1-7BModel size=7B2026.02 | 5,190 | — | — | — | — | — | — | 74.7 | 24.1 | 64.6 | 18.2 | 57.4 | 23.4 | 80.6 | 31.8 | 75.7 | 39.6 | 57 | 29.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InfiGUI-G1-7B2025.12 | 5,190 | — | — | — | — | — | — | 74.7 | 24.1 | 64.6 | 18.2 | 57.4 | 23.4 | 80.6 | 31.8 | 75.7 | 39.6 | 57 | 29.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-Venus-7BModel size=7B2026.02 | 5,080 | — | — | — | — | — | — | 74.7 | 24.1 | 63.1 | 14.7 | 60.4 | 21.9 | 76.4 | 31.8 | 75.7 | 41.5 | 49.5 | 22.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-Venus-7B2025.12 | 5,080 | — | — | — | — | — | — | 74.7 | 24.1 | 63.1 | 14.7 | 60.4 | 21.9 | 76.4 | 31.8 | 75.7 | 41.5 | 49.5 | 22.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTA1-7BModel size=7B2026.02 | 5,010 | — | — | — | — | — | — | 66.9 | 20.7 | 62.6 | 18.9 | 53.3 | 17.2 | 76.4 | 31.8 | 82.5 | 50.9 | 48.6 | 25.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTA1-7B2025.12 | 5,010 | — | — | — | — | — | — | 66.9 | 20.7 | 62.6 | 18.9 | 53.3 | 17.2 | 76.4 | 31.8 | 82.5 | 50.9 | 48.6 | 25.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenCUA-7BModel size=7B2026.02 | 5,000 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenCUA-7B2025.12 | 5,000 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-8BModel size=8B2026.02 | 4,990 | — | — | — | — | — | — | 79.2 | 23.4 | 68.2 | 14 | 46.7 | 10.9 | 73.6 | 30 | 76.3 | 30.2 | 65.4 | 21.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-8BEvaluated by us=true2025.12 | 4,990 | — | — | — | — | — | — | 79.2 | 23.4 | 68.2 | 14 | 46.7 | 10.9 | 73.6 | 30 | 76.3 | 30.2 | 65.4 | 21.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G2-7BModel size=7B2026.02 | 4,750 | — | — | — | — | — | — | 68.8 | 17.2 | 57.1 | 15.4 | 55.8 | 12.5 | 77.1 | 24.5 | 74 | 32.7 | 57.9 | 21.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G2-7B2025.12 | 4,750 | — | — | — | — | — | — | 68.8 | 17.2 | 57.1 | 15.4 | 55.8 | 12.5 | 77.1 | 24.5 | 74 | 32.7 | 57.9 | 21.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G2Evaluation Mode=Upper Bound2026.01 | 4,750 | — | — | — | — | — | — | 68.8 | 17.2 | 57.1 | 15.4 | 55.8 | 12.5 | 77.1 | 24.5 | 74 | 32.7 | 57.9 | 21.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SE-GUI-7BModel Type=Open-Source2025.12 | 4,730 | 4,450 | 3,720 | 42.1 | 54.7 | 7,040 | 3,880 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SE-GUI-7BModel size=7B2026.02 | 4,720 | — | — | — | — | — | — | 68.2 | 19.3 | 57.6 | 9.1 | 51.3 | 14.1 | 75 | 28.2 | 78.5 | 43.4 | 49.5 | 25.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SE-GUI-7B2025.12 | 4,720 | — | — | — | — | — | — | 68.2 | 19.3 | 57.6 | 9.1 | 51.3 | 14.1 | 75 | 28.2 | 78.5 | 43.4 | 49.5 | 25.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InfiGUI-3BModel size=3B2026.02 | 4,520 | — | — | — | — | — | — | 64.9 | 20 | 51.5 | 16.8 | 50.8 | 25 | 68.8 | 32.7 | 70.6 | 32.1 | 49.5 | 19.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| InfiGUI-3B2025.12 | 4,520 | — | — | — | — | — | — | 64.9 | 20 | 51.5 | 16.8 | 50.8 | 25 | 68.8 | 32.7 | 70.6 | 32.1 | 49.5 | 19.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Actor-7BModel size=7B2026.02 | 4,460 | — | — | — | — | — | — | 59.1 | 15.9 | 59.6 | 16.1 | 47.7 | 9.4 | 70.1 | 25.5 | 69.5 | 41.5 | 55.1 | 19.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-Actor-7B2025.12 | 4,460 | — | — | — | — | — | — | 59.1 | 15.9 | 59.6 | 16.1 | 47.7 | 9.4 | 70.1 | 25.5 | 69.5 | 41.5 | 55.1 | 19.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Phi-Ground2026.02 | 4,320 | — | — | — | — | — | — | 70.8 | 16.7 | 56.6 | 13.3 | 26.9 | 17.2 | 58 | 29.1 | 76.4 | 44 | 55.1 | 25.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Phi-Ground2025.12 | 4,320 | — | — | — | — | — | — | 70.8 | 16.7 | 56.6 | 13.3 | 26.9 | 17.2 | 58 | 29.1 | 76.4 | 44 | 55.1 | 25.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-1.5-7BModel Type=Open-Source2025.12 | 4,200 | 3,180 | 4,020 | 31.8 | 47.2 | 6,560 | 3,320 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-2BModel size=2B2026.02 | 4,190 | — | — | — | — | — | — | 55.2 | 11.7 | 59.1 | 16.1 | 31 | 15.6 | 64.6 | 22.7 | 72.3 | 34 | 59.8 | 23.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-VL-2BEvaluated by us=true2025.12 | 4,190 | — | — | — | — | — | — | 55.2 | 11.7 | 59.1 | 16.1 | 31 | 15.6 | 64.6 | 22.7 | 72.3 | 34 | 59.8 | 23.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Jedi-7BModel Type=Open-Source2025.12 | 3,950 | 2,740 | 3,400 | 32.2 | 52.4 | 6,870 | 2,600 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-72BModel Type=Open-Source2025.12 | 3,810 | 4,080 | 3,960 | 17.2 | 45.7 | 5,480 | 3,010 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-Tars-72BModel size=72B2026.02 | 3,810 | — | — | — | — | — | — | 63 | 17.2 | 57 | 15.4 | 18.8 | 12.5 | 64.6 | 20.9 | 63.3 | 26.4 | 42.1 | 15.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-72B2025.12 | 3,810 | — | — | — | — | — | — | 63 | 17.2 | 57 | 15.4 | 18.8 | 12.5 | 64.6 | 20.9 | 63.3 | 26.4 | 42.1 | 15.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G1-3BModel Type=Open-Source2025.12 | 3,710 | 3,110 | 2,660 | 32.2 | 48 | 5,910 | 1,610 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OperatorModel Type=Proprietary2025.12 | 3,660 | 3,510 | 3,960 | 16.1 | 43.7 | 5,300 | 3,270 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Jedi-3BModel Type=Open-Source2025.12 | 3,610 | 3,810 | 3,460 | 23 | 38.6 | 5,700 | 2,500 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SE-GUI-3BModel Type=Open-Source2025.12 | 3,590 | 3,510 | 2,900 | 31.8 | 43.3 | 5,090 | 2,550 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-7BModel size=7B2026.02 | 3,570 | — | — | — | — | — | — | 58.4 | 12.4 | 50 | 9.1 | 20.8 | 9.4 | 63.9 | 31.8 | 63.3 | 20.8 | 30.8 | 16.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-7B2025.12 | 3,570 | — | — | — | — | — | — | 58.4 | 12.4 | 50 | 9.1 | 20.8 | 9.4 | 63.9 | 31.8 | 63.3 | 20.8 | 30.8 | 16.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround-72BModel Type=Open-Source2025.12 | 3,450 | 3,110 | 3,580 | 13.8 | 50 | 5,130 | 2,550 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround-v1-72BModel size=72B2026.02 | 3,450 | — | — | — | — | — | — | 55.8 | 4.8 | 54 | 10.5 | 16.8 | 4.7 | 70.8 | 22.7 | 61 | 18.9 | 40.2 | 7.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround-v1-72B2025.12 | 3,450 | — | — | — | — | — | — | 55.8 | 4.8 | 54 | 10.5 | 16.8 | 4.7 | 70.8 | 22.7 | 61 | 18.9 | 40.2 | 7.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSL-7BProtocol=Reinforcement Learning, Paradigm=SSL, Training Data=Mix-3K2026.01 | 2,910 | — | — | — | — | — | — | 50.7 | 5.5 | 42 | 9.1 | 24.4 | 5.1 | 59 | 13.7 | 57.1 | 26.5 | 42.1 | 13.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RL-Continuous-7BProtocol=Reinforcement Learning, Paradigm=RLVR, Training Data=Mix-3K2026.01 | 2,850 | — | — | — | — | — | — | 48 | 4.1 | 40.9 | 9 | 27.4 | 4.7 | 61.1 | 12.7 | 56.4 | 24.5 | 41.1 | 12.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Claude 3.7 SonnetModel Type=Proprietary2025.12 | 2,770 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-7BModel Type=Open-Source2025.12 | 2,760 | 2,910 | 2,490 | 13.8 | 31.1 | 4,570 | 2,240 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RL-Binary-7BProtocol=Reinforcement Learning, Paradigm=RLVR, Training Data=Mix-3K2026.01 | 2,620 | — | — | — | — | — | — | 40 | 3.4 | 40.4 | 6.9 | 22.8 | 4.7 | 54.8 | 11.8 | 55.9 | 20.7 | 41.1 | 12.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSL-3BProtocol=Reinforcement Learning, Paradigm=SSL, Training Data=Mix-3K2026.01 | 2,470 | — | — | — | — | — | — | 31.4 | 4.9 | 37.9 | 7 | 26.4 | 10.9 | 61.8 | 21.9 | 45.8 | 13.2 | 27.1 | 7.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RL-Continuous-3BProtocol=Reinforcement Learning, Paradigm=RLVR, Training Data=Mix-3K2026.01 | 2,320 | — | — | — | — | — | — | 27.3 | 4.2 | 40.4 | 5.6 | 27.4 | 9.4 | 54.9 | 19.1 | 49.2 | 11.4 | 26.2 | 3.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RL-Binary-3BProtocol=Reinforcement Learning, Paradigm=RLVR, Training Data=Mix-3K2026.01 | 2,240 | — | — | — | — | — | — | 28 | 4.2 | 35.4 | 4.2 | 25.9 | 7.8 | 61.1 | 20 | 44.7 | 7.5 | 25.3 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-AiFResolution=Normal to High resolution shift2026.01 | 1,900 | — | — | — | — | — | — | 29.2 | 2.1 | 23.2 | 3.5 | 20.3 | 15.6 | 33.3 | 14.7 | 38.4 | 18.2 | 22.4 | 6.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OS-Atlas-7BModel Type=Open-Source2025.12 | 1,890 | 1,770 | 1,790 | 10.3 | 24.4 | 2,740 | 1,680 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QwenVL2.5-7B*Protocol=Post-Training, SFT=Mix-3K2026.01 | 1,840 | — | — | — | — | — | — | 31.4 | 1.8 | 27.3 | 3.5 | 15.7 | 5.1 | 40.7 | 7.9 | 39.7 | 8.9 | 32.4 | 6.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-AiFResolution=Normal2026.01 | 1,770 | — | — | — | — | — | — | 24.7 | 1.4 | 17.7 | 8.2 | 27.4 | 4.7 | 29.2 | 14.6 | 35.3 | 20.8 | 17.8 | 10.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QwenVL2.5-7BProtocol=Zero Shot2026.01 | 1,730 | — | — | — | — | — | — | 33.1 | 2.1 | 23.7 | 3.5 | 12.2 | 6.3 | 36.8 | 7.3 | 37.8 | 7.5 | 30.8 | 6.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Claude C.2026.02 | 1,710 | — | — | — | — | — | — | 22 | 3.9 | 25.9 | 3.4 | 14.5 | 3.7 | 33.9 | 15.8 | 30.1 | 16.3 | 11 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Claude C.Type=Proprietary2025.12 | 1,710 | — | — | — | — | — | — | 22 | 3.9 | 25.9 | 3.4 | 14.5 | 3.7 | 33.9 | 15.8 | 30.1 | 16.3 | 11 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Claude Computer Use2026.01 | 1,710 | — | — | — | — | — | — | 22 | 3.9 | 25.9 | 3.4 | 14.5 | 3.7 | 33.9 | 15.8 | 30.1 | 16.3 | 11 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SeeClickResolution=Normal to High resolution shift2026.01 | 1,670 | — | — | — | — | — | — | 25.3 | 0 | 20.8 | 5.6 | 26.4 | 6.3 | 28.1 | 14.6 | 35.6 | 11.3 | 17.7 | 8.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G2Resolution=Normal to High resolution shift2026.01 | 1,670 | — | — | — | — | — | — | 24.4 | 4.1 | 23.7 | 5.6 | 24.5 | 6.3 | 27.3 | 12.6 | 35.1 | 9.4 | 21.2 | 5.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Os-Atlas-7BProtocol=Post-Training2026.01 | 1,660 | — | — | — | — | — | — | 33.1 | 1.4 | 28.8 | 2.8 | 12.2 | 4.7 | 37.5 | 7.3 | 33.9 | 5.7 | 27.1 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5VL-3B2026.01 | 1,610 | — | — | — | — | — | — | 22.1 | 1.4 | 26.8 | 2.1 | 9.1 | 7.3 | 38.2 | 7.3 | 33.9 | 15.1 | 10.3 | 1.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-R1-3BProtocol=Post-Training2026.01 | 1,600 | — | — | — | — | — | — | 22.7 | 4.1 | 27.3 | 3.5 | 11.2 | 6.3 | 43.4 | 11.8 | 32.2 | 11.3 | 13.1 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Claude**Protocol=Post-Training2026.01 | 1,540 | — | — | — | — | — | — | 22 | 3.9 | 25.9 | 3.4 | 14.5 | 3.7 | 33.9 | 15.8 | 30.1 | 16.3 | 11 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G2Resolution=Normal2026.01 | 1,470 | — | — | — | — | — | — | 17.5 | 2.1 | 20.2 | 6.3 | 24.3 | 1.7 | 22.2 | 12.8 | 36.2 | 11.3 | 17.7 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround-7BProtocol=Post-Training2026.01 | 1,420 | — | — | — | — | — | — | 26.6 | 2.1 | 27.3 | 2.8 | 14.2 | 1.6 | 31.9 | 2.7 | 31.6 | 11.3 | 17.8 | 0 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QwenVL2.5-3B*Protocol=Post-Training, SFT=Mix-3K2026.01 | 1,380 | — | — | — | — | — | — | 20.3 | 1.8 | 24.6 | 2.8 | 11.2 | 4.7 | 39.5 | 6.4 | 28.6 | 5.7 | 17.8 | 2.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| QwenVL2.5-3BProtocol=Zero Shot2026.01 | 1,220 | — | — | — | — | — | — | 16.2 | 1.4 | 23.3 | 1.4 | 10.2 | 4.7 | 38.2 | 6.4 | 24.3 | 3.8 | 15 | 1.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FOCUS-2BProtocol=Post-Training2026.01 | 1,200 | — | — | — | — | — | — | 22.8 | 1.7 | 23.7 | 1.7 | 7.6 | 3.1 | 25 | 7.1 | 23.2 | 7.7 | 17.8 | 2.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |