GUI Grounding on ScreenSpot Pro
72.7AccuracyGemini-3-Pro
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Gemini-3-ProAccess=API call only2026.03 | 72.7 | — | — | — | — | |
| vLLMBase Model=GUI-Owl-1.5-32B-Instruct2026.05 | 70.97 | — | — | — | 3.3 | |
| AQuaUIBase Model=GUI-Owl-1.5-32B-Instruct2026.05 | 69.89 | — | — | — | 2.77 | |
| UI-Venus-1.5-30B-A3Bdesign=end-to-end2026.02 | 69.6 | — | — | — | — | |
| vLLMBase Model=GUI-Owl-1.5-8B-Instruct2026.05 | 68.88 | — | — | — | 2.69 | |
| UI-Venus-1.5-8Bdesign=end-to-end2026.02 | 68.4 | — | — | — | — | |
| UI-Venus-1.5-8BAccess=Open weights2026.03 | 68.4 | — | — | — | — | |
| quality-aware self-distillation2026.06 | 68.37 | — | — | — | — | |
| MAI-UI-32Bdesign=end-to-end2026.02 | 67.9 | — | — | — | — | |
| AQuaUIBase Model=GUI-Owl-1.5-8B-Instruct2026.05 | 67.68 | — | — | — | 2.21 | |
| Naive-OPSD2026.06 | 66.41 | — | — | — | — | |
| ZoomOnceSize=4B2026.06 | 66.2 | — | — | — | — | |
| Holo2-30B-A3Bdesign=end-to-end2026.02 | 66.1 | — | — | — | — | |
| MAI-UI-8Bdesign=end-to-end2026.02 | 65.8 | — | — | — | — | |
| MAI-UI-8BAccess=Open weights2026.03 | 65.8 | — | — | — | — | |
| SFT2026.06 | 65.15 | — | — | — | — | |
| MAI-UISize=4B2026.06 | 65.1 | — | — | — | — | |
| GUI-SD2026.06 | 64.71 | — | — | — | — | |
| vLLMBase Model=MAI-UI-8B2026.05 | 64.52 | — | — | — | 2.15 | |
| Seed1.8design=end-to-end2026.02 | 64.3 | — | — | — | — | |
| GTA1-32Bdesign=end-to-end2026.02 | 63.6 | — | — | — | — | |
| GRPO2026.06 | 63.5 | — | — | — | — | |
| EvoCUA-OpenCUA-72BBase Model=OpenCUA-72B2026.01 | 63.24 | — | — | — | — | |
| Qwen3.5-9BParameters=9B2026.06 | 63 | — | — | — | — | |
| Step-GUI-8Bdesign=end-to-end2026.02 | 62.6 | — | — | — | — | |
| Step-GUI-8BParameters=8B2025.12 | 62.6 | — | — | — | — | |
| UI-Venus-1.0-72Bdesign=end-to-end2026.02 | 61.9 | — | — | — | — | |
| UI-TARS 1.52025.05 | 61.6 | — | — | — | — | |
| UI-TARS-1.5Version=1.52025.12 | 61.6 | — | — | — | — | |
| UI-TARS 1.5-7BAccess=Open weights2026.03 | 61.6 | — | — | — | — | |
| Vero Q3I-8BRL=Qwen3VL, Initial Model=8B Inst2026.04 | 61.4 | — | — | — | — | |
| MolmoPoint-GUI-8BGUI-specific Fine-tuning=true, Test-time scaling=64crops2026.03 | 61.1 | — | — | — | — | |
| Seed 1.5-VL2025.05 | 60.9 | — | — | — | — | |
| SeedVL-1.5Version=1.52025.12 | 60.9 | — | — | — | — | |
| OpenCUA-72BParameters=72B2026.01 | 60.8 | — | — | — | — | |
| OpenCUA-72Bdesign=end-to-end2026.02 | 60.8 | — | — | — | — | |
| GUI-SDTime/epoch=4.22026.05 | 60.7 | — | — | — | — | |
| Qwen3-30B-A3B-InstructParameters=30B, Arch=A3B, Type=Instruct2025.12 | 60.5 | — | — | — | — | |
| MolmoPoint-GUI-8BGUI-specific Fine-tuning=true2026.03 | 60.2 | — | — | — | — | |
| Step-GUI-4Bdesign=end-to-end2026.02 | 60 | — | — | — | — | |
| Step-GUI-4BParameters=4B2025.12 | 60 | — | — | — | — | |
| Qwen3-VL-4B-InstructParameters=4B, Type=Instruct2025.12 | 59.5 | — | — | — | — | |
| AQuaUIBase Model=MAI-UI-8B2026.05 | 59.14 | — | — | — | 1.88 | |
| Holo2-8Bdesign=end-to-end2026.02 | 58.9 | — | — | — | — | |
| Holo2-8BAccess=Open weights2026.03 | 58.9 | — | — | — | — | |
| vLLMBase Model=MAI-UI-2B2026.05 | 58.89 | — | — | — | 1.25 | |
| Propose-then-Critic-8B2026.05 | 58.7 | — | — | — | — | |
| vLLMBase Model=Qwen3-VL-32B-Instruct2026.05 | 58.57 | — | — | — | 3.24 | |
| Image-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:5, Avg #Image Tokens=4,195, FLOPs (G)=75.82025.12 | 58.35 | — | — | — | — | |
| Std-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:5, Avg #Image Tokens=4,195, FLOPs (G)=496.92025.12 | 58.18 | — | — | — | — | |
| GUI-Owl-32Bdesign=end-to-end2026.02 | 58 | — | — | — | — | |
| GUI-Owl-32BParameters=32B2025.12 | 58 | — | — | — | — | |
| GUI-OwlSize=32B2026.06 | 58 | — | — | — | — | |
| UI-Venus-1.5-2Bdesign=end-to-end2026.02 | 57.7 | — | — | — | — | |
| AQuaUIBase Model=MAI-UI-2B2026.05 | 57.5 | — | — | — | 1.25 | |
| MAI-UI-2Bdesign=end-to-end2026.02 | 57.4 | — | — | — | — | |
| Qwen3-VL-Instruct + GRPO-GaussianTime/epoch=16.82026.05 | 57.4 | — | — | — | — | |
| Qwen3-VL-32B-ThinkingParameters=32B, Thinking Mode=true2026.01 | 57.1 | — | — | — | — | |
| UI-InsSize=7B2026.06 | 57 | — | — | — | — | |
| ZwZ-8B2026.05 | 56.8 | — | — | — | — | |
| Qwen3-VL-Instruct + GRPO-BinaryTime/epoch=16.92026.05 | 56.8 | — | — | — | — | |
| Qwen3-VL-Instruct + GRPO-DistanceTime/epoch=16.72026.05 | 56.6 | — | — | — | — | |
| vLLMBase Model=Qwen3-VL-8B-Instruct2026.05 | 56.23 | — | — | — | 1.95 | |
| AQuaUIBase Model=Qwen3-VL-32B-Instruct2026.05 | 56.17 | — | — | — | 2.44 | |
| Image-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:4, Avg #Image Tokens=3,367, FLOPs (G)=59.492025.12 | 55.94 | — | — | — | — | |
| OpenCUA-32Bdesign=end-to-end2026.02 | 55.3 | — | — | — | — | |
| OpenCUASize=32B2026.06 | 55.3 | — | — | — | — | |
| AQuaUIBase Model=Qwen3-VL-8B-Instruct2026.05 | 55.28 | — | — | — | 1.67 | |
| GUI-Owl-7Bdesign=end-to-end2026.02 | 54.9 | — | — | — | — | |
| GUI-Owl-7BParameters=7B2025.12 | 54.9 | — | — | — | — | |
| GUI-OwlSize=7B2026.06 | 54.9 | — | — | — | — | |
| Qwen3-VLSize=32B2026.06 | 54.9 | — | — | — | — | |
| Qwen3-VL-8B-InstructParameters=8B, Type=Instruct2025.12 | 54.6 | — | — | — | — | |
| Std-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:4, Avg #Image Tokens=3,367, FLOPs (G)=415.52025.12 | 54.04 | — | — | — | — | |
| Qwen3-VL-30B-A3Breproduction=true, design=end-to-end2026.02 | 53.7 | — | — | — | — | |
| Qwen3-VL-Instruct2026.05 | 53.6 | — | — | — | — | |
| Ferret-UI LiteSize=–2026.06 | 53.3 | — | — | — | — | |
| ZoomUI-7B2026.05 | 52.8 | — | — | — | — | |
| Qwen3-VL-8Breproduction=true, design=end-to-end2026.02 | 52.7 | — | — | — | — | |
| Qwen3-VL-8BAccess=Open weights2026.03 | 52.7 | — | — | — | — | |
| Molmo2-GUI-8BAccess=Fully open, GUI-specific Fine-tuning=true2026.03 | 52.3 | — | — | — | — | |
| UI-Ins-7B2026.05 | 52.2 | — | — | — | — | |
| InfiGUI-G1Size=7B2026.06 | 51.9 | — | — | — | — | |
| UI-Venus-1.0-7Bdesign=end-to-end2026.02 | 50.8 | — | — | — | — | |
| UI-VenusSize=7B2026.06 | 50.8 | — | — | — | — | |
| Std-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:3, Avg #Image Tokens=2,534, FLOPs (G)=333.72025.12 | 50.6 | — | — | — | — | |
| JEDI-7BAccess=Fully open2026.03 | 50.2 | — | — | — | — | |
| GroundCUA-7BAccess=Fully open2026.03 | 50.2 | — | — | — | — | |
| GTA1-7Bdesign=end-to-end2026.02 | 50.1 | — | — | — | — | |
| GTA1-7BAccess=Fully open2026.03 | 50.1 | — | — | — | — | |
| GTA1-7B2026.05 | 50.1 | — | — | — | — | |
| GTA1Size=7B2026.06 | 50.1 | — | — | — | — | |
| Image-LoRABackbone=QWEN2.5-VL-72B, Text:Img Token Ratio=1:3, Avg #Image Tokens=2,534, FLOPs (G)=43.82025.12 | 50.09 | — | — | — | — | |
| OpenCUA-7Bdesign=end-to-end2026.02 | 50 | — | — | — | — | |
| OpenCUA-7BAccess=Fully open2026.03 | 50 | — | — | — | — | |
| OpenCUASize=7B2026.06 | 50 | — | — | — | — | |
| EvoCUA-32BParameters=32B2026.01 | 49.76 | — | — | — | — | |
| Qwen3-VLSize=8B2026.06 | 49.5 | — | — | — | — | |
| HyperClick-7B2026.05 | 48.2 | — | — | — | — | |
| GPT-4o + OS-Atlas-7BType=Planner+Grounder, Params=≫7B2026.03 | 48.1 | — | — | — | — |