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

Visual Reasoning on MMMU-Pro

39.42Avg@8

Vanilla

10.081617.698325.31532.9317Mar 24, 2026Mar 27, 2026Mar 31, 2026Apr 3, 2026Apr 7, 2026Apr 10, 2026Apr 14, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.04
39.42-100
2026.04
37.63-100
2026.04
34.2786.992.44
2026.04
31.2179.290.09
2026.04
31.0478.787
2026.04
30.5281.182.9
2026.04
28.2671.787.83
2026.04
28.2171.682.98
2026.04
28.0374.583.8
2026.03
27.47--
2026.04
27.3972.881.04
2026.03
27.11--
2026.04
26.8771.471.48
2026.04
25.8565.681.01
2026.03
25.66--
2026.04
22.2559.168.74
2026.04
21.6657.673.34
2026.04
19.4249.363.55
2026.04
19.4151.655.68
2026.04
18.9648.156.66
2026.04
16.7142.459.54
2026.04
16.3643.560.42
2026.04
14.135.857.61
2026.04
13.0834.855.24
2026.04
1334.546.4
2026.04
12.0230.551.04
2026.04
11.629.449.42
2026.04
11.530.645.18
2026.04
11.2129.847.4