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

Agentic Reasoning on AndroidWorld

82.05Success Rate

EAPO-8B

31.079644.312357.54570.7777May 9, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.05
82.05
2026.05
79.59
76.36
2026.05
73.38
2026.05
71.6
2026.05
68.48
2026.05
62
2026.05
61.76
2026.05
61.75
2026.05
60.98
2026.05
58.87
2026.05
57.99
2026.05
55.48
2026.05
55.43
2026.05
52.04
2026.05
51.66
2026.05
49.31
2026.05
48.93
2026.05
47.52
2026.05
33.04