List your company

Operating System GUI Agentic Reasoning on OSWorld

64.29Success RateEAPO-8B
18.436430.340742.24554.1493May 9, 2026

Evaluation results

MethodLinks
EAPO-8BModel Category=Ours2026.0564.29
EAPO-8B2026.0564.29
Claude-4-Sonnet-0929Model Category=Closed-source Models2026.0562.88
Claude-4-Sonnet-0929Category=Closed-source2026.0562.88
EAPO-4BModel Category=Ours2026.0557.89
EAPO-4B2026.0557.89
LAMERCategory=Training Method2026.0555.6
DeepMiner-Mano-72BModel Category=Closed-source Models2026.0553.88
DeepMiner-Mano-72BCategory=Closed-source2026.0553.88
UI-TARS-2-2509Model Category=Closed-source Models2026.0553.11
UI-TARS-2-2509Category=Closed-source2026.0553.11
GiGPOCategory=Training Method2026.0550.88
EAPO-1.7B/2BModel Category=Ours2026.0550.34
EAPO-1.7B/2B2026.0550.34
DAPOCategory=Training Method2026.0547.9
DART-GUI-7BModel Category=Open-source Models2026.0542.13
DART-GUI-7BCategory=Open-source2026.0542.13
Qwen3-VL-32BModel Category=Open-source Models2026.0541
GRPOCategory=Training Method2026.0540.36
Seed1.5-VL-250717Model Category=Closed-source Models2026.0540.18
Seed1.5-VL-250717Category=Closed-source2026.0540.18
DeepMiner-Mano-7BModel Category=Closed-source Models2026.0540.16
Qwen3-VL-235B-A22BModel Category=Open-source Models2026.0538.1
Qwen3-VL-235B-A22BCategory=Open-source2026.0538.1
Claude-3.7-SonnetModel Category=Closed-source Models2026.0535.83
OpenCUA-32BModel Category=Open-source Models2026.0534.79
GUI-Owl-7BModel Category=Open-source Models2026.0534.79
OpenCUA-32BCategory=Open-source2026.0534.79
GUI-Owl-7BCategory=Open-source2026.0534.79
OpenAI CUAModel Category=Closed-source Models2026.0531.3
ARPOModel Category=Open-source Models2026.0529.9
ARPOCategory=Open-source2026.0529.9
TianXi-Action-7BModel Category=Closed-source Models2026.0529.81
TianXi-Action-7BCategory=Closed-source2026.0529.81
OpenCUA-7BModel Category=Open-source Models2026.0528.2
UI-TARS-7BModel Category=Open-source Models2026.0527.52
UI-TARS-7BCategory=Open-source2026.0527.52
UI-TARS-72B-dpoModel Category=Open-source Models2026.0526.84
OpenAI CUA o3Model Category=Closed-source Models2026.0523
OpenAI CUA o3Category=Closed-source2026.0523
ZeroGUIModel Category=Open-source Models2026.0520.2
ZeroGUICategory=Open-source2026.0520.2