List your company

Mobile Automation on Android In The Wild (AITW)

7.7Average ScoreChatGPT-COT
4.895223.827642.7661.6924Jul 8, 2025Aug 5, 2025Sep 3, 2025Oct 1, 2025Oct 30, 2025Nov 27, 2025Dec 26, 2025

Evaluation results

MethodLinks
ChatGPT-COTprompting=Chain-of-Thought2025.077.75.94.410.59.48.4———————
LLaMa 2Base Model=LLaMa 2, Size=7B, Vision Only=false2025.1228.428.5635.1830.9927.3519.92———————
PaLM2-COTprompting=Chain-of-Thought2025.0739.6————————————
GPT-4Vcategory=General-purpose VLM2025.0750.541.742.649.872.845.7———————
Qwen-VL2025.0754.349.559.946.964.750.757.4——————
GPT-4oVision Only=true2025.1255.0247.0649.1252.380.2846.42———————
OmniParserVision Only=true2025.1257.648.357.851.677.452.9———————
SeeClickmodel_size=9.6B2025.0759.35466.454.963.557.666.4——————
R-VLMmode=fine-tuned2025.0764.959.970.659.672.561.771——————
MobileAgentBase Model=LLaMa, Size=7B, Vision Only=false2025.1266.9255.874.9863.9576.2763.61———————
Qwen2 VLBase Model=Qwen2 VL, Size=2B, Vision Only=true2025.1267.2461.471.862.673.766.7———————
ShowUIBase Model=Qwen2 VL, Size=2B, Vision Only=true2025.1270.0463.972.569.777.566.6———————
TongUIBase Model=Qwen2.5 VL, Size=3B, Vision Only=true2025.1271.665.675.174.57765.8———————
TongUIBase Model=Qwen2.5 VL, Size=7B, Vision Only=true2025.1273.367.676.373.579.969.1———————
AutoGUIBase Model=T5 + ViT, Size=4.5B, Vision Only=true2025.1274.2768.2476.8971.3784.5870.26———————
Mobile VLMBase Model=QwenVL, Size=9.6B, Vision Only=true2025.1275.4269.5879.8774.7281.2471.7———————
SeeClickBase Model=QwenVL, Size=9.6B, Vision Only=true2025.1276.1667.679.675.984.673.1———————
SphAgentBase Model=InternLM, Size=7B, Vision Only=true2025.1276.2868.280.573.385.474———————
iSHIFTBase Model=Qwen2 VL, Size=2.5B, Vision Only=true2025.1276.3470.680.8271.6486.0372.6———————
CogAgentBase Model=Vicuna 7B, Size=18B, Vision Only=true2025.1276.8865.3878.8674.9593.4971.73———————
CoCo-AgentBase Model=LLaMa 2, Size=7B, Vision Only=false2025.1277.8269.9280.675.7688.8174.02———————
Auto-UI2024.01———————68.276.971.484.670.374.3
CogAgent2024.01———————65.478.97593.571.176.9
SeeClick2024.01———————67.679.675.984.673.176.2