List your company

Agentic Performance on OSWorld Verified

70.8OS PerformanceClaude-4.5-Sonnet
5.822.67539.5556.425Nov 10, 2025

Evaluation results

MethodLinks
Claude-4.5-SonnetModel Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1170.872.661.463.34962.9
OpenAI o3Model Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1162.514.521.438.816.523
GROUNDNEXT-3BModel Category=Open-source Models, Training Protocol=RL, Planner=OpenAI o3, Resolution=1920x10802025.1162.5475573.536.550.6
OpenCUA-72BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1158.34753.873.520.446.1
JEDI-7BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.115046.161.975.535.351
Claude-4-SonnetModel Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1145.839.348.159.227.941.4
UI-TARS-250705Model Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1141.750.455.75114.741.8
OpenCUA-7BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1141.722.237.1499.327
Qwen3-VL-FlashModel Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1140.953.655.1222241.6
UI-TARS-72B-DPOModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1137.51934.663.38.327.1
UI-TARS-1.5-7BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1133.329.937.953.19.129.6
CUAModel Category=Proprietary Models, Planner=OpenAI o3, Resolution=1920x10802025.1123.934.655.118.318.331.4
Qwen2.5-VL-72BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1116.74.36.423.25
Kimi-VL-A3BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1112.5621.718.41.110.3
OpenCUA-A3BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.1112.516.321.746.92.217.7
Qwen2.5-VL-32BModel Category=Open-source Models, Planner=OpenAI o3, Resolution=1920x10802025.118.31.76.46.12.23.9