Long-horizon GUI Interaction on MemGUI-Bench benchmark leaderboard · SOTA2 Research

Long-horizon GUI Interaction on MemGUI-Bench

50Precision@1 (1 App)Agent-S2

Evaluation Results

Method	Links
Agent-S22026.04		50	78.6	19.6	35.7	26.5	52.9	10	30	41.7	64.6	19	42.9	18.4	36.8	27.3	49.2
M3A2026.04		46.4	64.3	28.6	41.1	29.4	44.1	30	50	39.6	47.9	35.7	50	21.1	44.7	32.8	47.7
UI-Copilot-7B*2026.04		42.9	50	12.5	16.1	5.9	8.8	0	10	29.2	33.3	13.2	18.4	4.8	7.1	16.4	20.3
T3A2026.04		42.9	60.7	16.1	37.5	23.5	38.2	0	30	31.2	45.8	16.7	45.2	18.4	34.2	22.7	42.2
Mobile-Agent-E2026.04		25	42.9	0	1.8	0	0	0	0	12.5	22.9	2.4	2.4	0	2.6	5.5	10.2
UI-Venus-7B2026.04		21.4	28.6	1.8	1.8	0	2.9	0	0	14.6	20.8	0	0	0	0	5.5	7.8
GUI-Owl-7B2026.04		21.4	35.7	1.8	1.8	2.9	5.9	0	0	14.6	22.9	0	2.4	2.6	2.6	6.2	10.2
UI-S1-7B*2026.04		17.9	21.4	3.6	3.6	0	0	0	0	12.5	14.6	2.6	2.6	0	0	5.5	6.2
Qwen2.5VL-7B*2026.04		14.3	17.9	1.8	1.8	0	0	0	0	10.4	12.5	0	0	0	0	3.9	4.7
UI-TARS-1.5-7B2026.04		14.3	21.4	0	1.8	0	2.9	0	0	8.3	16.7	0	0	0	0	3.1	6.2
Mobile-Agent-V22026.04		14.3	17.9	0	0	0	0	0	0	8.3	10.4	0	0	0	0	3.1	3.9
AppAgent2026.04		14.3	42.9	0	0	0	0	0	0	8.3	22.9	0	2.4	0	0	3.1	9.4
SeeAct2026.04		10.7	25	0	0	0	0	0	0	6.2	12.5	0	2.4	0	0	2.3	5.5
CogAgent2026.04		0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0