List your company

Action Prediction on AndroidControl High v2

68.59Pass@1 Step AccuracyMem-W-8B
35.944444.419752.89561.3703May 10, 2026

Evaluation results

MethodLinks
Mem-W-8BBackbone=UI-Venus-1.5-8B, Memory Augmentation=Mem-W2026.0568.5981.1671.380.488.9482.96
GUI-Libra-8BModel Category=Open-source2026.0564.3——70.6——
Mem-W-4BBackbone=Qwen3-VL-4B, Memory Augmentation=Mem-W2026.0563.0774.6270.8574.8783.9280.72
GPT-5 + UGroundModel Category=Proprietary, Grounding Module=UGround2026.0561.3——69.4——
UI-Venus-1.5-8BModel Category=Baseline2026.0561.0674.6261.0670.180.482.51
Step-GUI-4BModel Category=Open-source2026.0559.374.3769.9671.3683.1780.72
GUI-Libra-7BModel Category=Open-source2026.0559.3——67.3——
Qwen3-VL-32BModel Category=Open-source2026.0559.0574.6269.9670.182.1677.58
GPT-4.1 + UGroundModel Category=Proprietary, Grounding Module=UGround2026.0557.5——63.3——
GPT-4o + UGroundModel Category=Proprietary, Grounding Module=UGround2026.0557——66.3——
Qwen3-VL-8BModel Category=Open-source2026.0554.869.3567.7166.177.8979.82
GPT-5-mini + UGroundModel Category=Proprietary, Grounding Module=UGround2026.0552.8——58.8——
UI-Venus-1.5-30B-A3BModel Category=Open-source2026.0551.2667.8464.1364.0774.6275.34
Qwen3-VL-4BModel Category=Baseline2026.0549.363.5769.0663.375.8877.13
Qwen3-VL-2BModel Category=Open-source2026.0545.4864.0759.6459.873.6273.99
UI-TARS-1.5-7BModel Category=Open-source2026.0545.270.657.463.184.1772.65
MAI-UI-2BModel Category=Open-source2026.0542.7161.0666.3743.4765.3362.78
GUI-Owl-1.5-8BModel Category=Open-source2026.0540.4559.0560.0952.7669.673.09
GUI-R1-3BModel Category=Open-source2026.0540——54——
GUI-R1-7BModel Category=Open-source2026.0539.7——56.3——
Aguvis-7BModel Category=Open-source2026.0537.7——43.7——
GLM-4.1V-9B-ThinkingModel Category=Open-source2026.0537.267.8448.434974.8759.64