List your company

End-to-end GUI Navigation on AndroidWorld

77.6Success RateUI-Venus-1.5-30B-A3B
28.09640.94853.866.652Feb 9, 2026Mar 4, 2026Mar 28, 2026Apr 21, 2026May 14, 2026Jun 7, 2026Jul 1, 2026

Evaluation results

MethodLinks
UI-Venus-1.5-30B-A3BParams.=30B2026.0277.6——
UI-Venus-1.5-8BParams.=8B2026.0273.7——
MAI-UI-32BParams.=32B2026.0273.3——
Holo2-30B-A3BParams.=30B2026.0271.6——
Seed1.82026.0270.7——
MAI-UI-8BParams.=8B2026.0270.7——
Gemini-2.5-Pro2026.0269.7——
Step-GUI-8BParams.=8B2026.0267.7——
GUI-Owl-7BParams.=7B2026.0266.4——
UI-Venus-1.0-72BParams.=72B2026.0265.9——
Step-GUI-4BParams.=4B2026.0263.9——
Qwen3.5-plusContext=Baseline2026.0761.2112.8—
Holo2-8BParams.=8B2026.0260.4——
Task-State Representation (Qwen3.5-plus)Base Model=Qwen3.5-plus, Context=Proposed2026.0757.7613.9-3.45
GLM-4.6VParams.=106B2026.0257——
UI-Venus-1.5-2BParams.=2B2026.0255.6——
Task-State Representation (Kimi-k2.5)Base Model=Kimi-k2.5, Context=Proposed2026.0755.1715.83.45
Qwen3-VL-30B-A3BParams.=30B2026.0254.3——
Kimi-k2.5Context=Baseline2026.0751.7215.5—
UI-Venus-1.0-7BParams.=7B2026.0249.1——
MAI-UI-2BParams.=2B2026.0249.1——
Qwen3-VL-8BParams.=8B2026.0247.6——
UI-TARS-72BParams.=72B2026.0246.6——
Qwen3-VL-2BParams.=2B2026.0236.4——
UI-TARS-1.5-7BParams.=7B2026.0230——