List your company

GUI Agent Task Planning on AITW

36.34General Step AccuracyGemini-2.0-flash-exp + Qwen2-VL-7B (AutoGUI)
13.990419.792725.59531.3973Feb 4, 2025

Evaluation results

MethodLinks
Gemini-2.0-flash-exp + Qwen2-VL-7B (AutoGUI)Planner=Gemini-2.0-flash-exp, Grounding Model=Qwen2-VL-7B SFT w/ AutoGUI2025.0236.3436.5450.9548.9540.9940.5250.9548.9532.8343.5239.23
Gemini-2.0-flash-expPlanner=Gemini-2.0-flash-exp, Grounding Model=Default/Proprietary2025.0226.3718.1628.4926.9130.322.8841.9428.9520.2222.6529.5
GPT-4o-mini + Qwen2-VL-7B (AutoGUI)Planner=GPT-4o-mini, Grounding Model=Qwen2-VL-7B SFT w/ AutoGUI2025.0220.4320.5625.5922.4915.2512.3325.5922.4916.1520.5318.37
GPT-4o-miniPlanner=GPT-4o-mini, Grounding Model=Default/Proprietary2025.0214.859.5811.175.7612.086.8521.0911.2410.8911.2214.01