Share your thoughts, 1 month free Claude Pro on usSee more

Operating System GUI Agentic Reasoning on OSWorld

64.29Success Rate

EAPO-8B

Updated 22d ago

Evaluation Results

Method	Links
EAPO-8B 2026.05		64.29
EAPO-8B 2026.05		64.29
Claude-4-Sonnet-0929 2026.05		62.88
Claude-4-Sonnet-0929 2026.05		62.88
EAPO-4B 2026.05		57.89
EAPO-4B 2026.05		57.89
LAMER 2026.05		55.6
DeepMiner-Mano-72B 2026.05		53.88
DeepMiner-Mano-72B 2026.05		53.88
UI-TARS-2-2509 2026.05		53.11
UI-TARS-2-2509 2026.05		53.11
GiGPO 2026.05		50.88
EAPO-1.7B/2B 2026.05		50.34
EAPO-1.7B/2B 2026.05		50.34
DAPO 2026.05		47.9
DART-GUI-7B 2026.05		42.13
DART-GUI-7B 2026.05		42.13
Qwen3-VL-32B 2026.05		41
GRPO 2026.05		40.36
Seed1.5-VL-250717 2026.05		40.18
Seed1.5-VL-250717 2026.05		40.18
DeepMiner-Mano-7B 2026.05		40.16
Qwen3-VL-235B-A22B 2026.05		38.1
Qwen3-VL-235B-A22B 2026.05		38.1
Claude-3.7-Sonnet 2026.05		35.83
OpenCUA-32B 2026.05		34.79
GUI-Owl-7B 2026.05		34.79
OpenCUA-32B 2026.05		34.79
GUI-Owl-7B 2026.05		34.79
OpenAI CUA 2026.05		31.3
ARPO 2026.05		29.9
ARPO 2026.05		29.9
TianXi-Action-7B 2026.05		29.81
TianXi-Action-7B 2026.05		29.81
OpenCUA-7B 2026.05		28.2
UI-TARS-7B 2026.05		27.52
UI-TARS-7B 2026.05		27.52
UI-TARS-72B-dpo 2026.05		26.84
OpenAI CUA o3 2026.05		23
OpenAI CUA o3 2026.05		23
ZeroGUI 2026.05		20.2
ZeroGUI 2026.05		20.2