Android GUI Automation on AndroidWorld M
53.8L1 ScoreStamp-GUI
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Stamp-GUIcategory=Ours (Training from GUI-Owl-1.5-8B-Think), pass@k=pass@32026.05 | 53.8 | 38.4 | 23.1 | 38.4 | |
| Claude-Sonnet-4.6category=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@12026.05 | 46.2 | 46.2 | 38.5 | 43.6 | |
| Claude-Sonnet-4.6category=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@32026.05 | 46.2 | 53.8 | 46.2 | 48.7 | |
| Stamp-GUIcategory=Ours (Training from GUI-Owl-1.5-8B-Think), pass@k=pass@12026.05 | 46.2 | 30.8 | 15.4 | 30.8 | |
| Gemini-3.1-Procategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@32026.05 | 38.4 | 23.1 | 23.1 | 28.2 | |
| Qwen3-VL-8B-Thinkcategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@32026.05 | 30.8 | 23.1 | 7.7 | 20.5 | |
| Gemini-3.1-Procategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@12026.05 | 30.8 | 15.4 | 23.1 | 23.1 | |
| Qwen3-VL-8B-Instructcategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@32026.05 | 23.1 | 15.4 | 7.7 | 15.4 | |
| Qwen3-VL-8B-Thinkcategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@12026.05 | 23.1 | 15.4 | 0 | 12.8 | |
| GPT-5.4category=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@12026.05 | 23.1 | 15.4 | 7.7 | 15.4 | |
| GPT-5.4category=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@32026.05 | 23.1 | 15.4 | 7.7 | 15.4 | |
| GUI-Owl-1.5-8B-Thinkcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@32026.05 | 23.1 | 15.4 | 7.7 | 15.4 | |
| Qwen3-VL-8B-Instructcategory=General-purpose Multimodal Model + Active Memory Instruction (AMI), pass@k=pass@12026.05 | 15.4 | 7.7 | 0 | 7.7 | |
| UI-Venus-1.5-8Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@32026.05 | 15.4 | 7.7 | 7.7 | 10.3 | |
| MAI-UI-8Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@12026.05 | 15.4 | 7.7 | 7.7 | 10.3 | |
| MAI-UI-8Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@32026.05 | 15.4 | 15.4 | 7.7 | 12.8 | |
| GUI-Owl-1.5-8B-Thinkcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@12026.05 | 15.4 | 7.7 | 7.7 | 10.3 | |
| CogAgent-9Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@32026.05 | 7.7 | 0 | 0 | 2.6 | |
| UI-Tars-1.5-7Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@12026.05 | 7.7 | 7.7 | 7.7 | 7.7 | |
| UI-Tars-1.5-7Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@32026.05 | 7.7 | 15.4 | 15.4 | 12.8 | |
| UI-Venus-1.5-8Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@12026.05 | 7.7 | 0 | 0 | 2.6 | |
| CogAgent-9Bcategory=GUI-specialized Model + History-Retention Protocol (HRP), pass@k=pass@12026.05 | 0 | 0 | 0 | 0 |