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

Multimodal Capability Evaluation on MMStar

68.3Overall Score

Baseline

31.69241.19650.760.204May 19, 2026May 25, 2026May 31, 2026Jun 6, 2026Jun 12, 2026Jun 18, 2026Jun 25, 2026
Updated 29d ago

Evaluation Results

MethodLinks
2026.06
68.3------74.9100
2026.05
6876.861.270.873.272.453.6--
2026.05
65.774.86069.666.867.655.6--
2026.06
65.5------73.1100
2026.05
64.17257.670.467.667.249.6--
2026.06
62.8------72.496.7
2026.05
61.570.448.86863.267.251.6--
2026.06
61.2------71.295.1
2026.05
60.77256.46861.658.448--
2026.06
59.5------70.496.3
2026.06
58.9------69.392.5
2026.06
57.9------68.894.1
2026.06
57.7------68.994.3
2026.06
57.1------68.891.9
2026.05
5669.644.867.259.260.834.4--
2026.06
55.7------67.490
2026.06
55.5------67.892.7
2026.06
55.1------67.890.5
2026.06
53.9------67.992.9
2026.06
53.8------65.589.6
2026.06
52.3------65.589.6
2026.06
52------64.686.2
2026.06
51.5------62.883.8
2026.05
50.455.635.661.654.456.438.8--
2026.05
49.86438.860.842.849.243.2--
2026.05
49.86646.457.650.845.232.8--
2026.06
49.2------60.380.5
2026.05
47.5605058.443.240.832.8--
2026.06
47.3------60.480.6
2026.06
43.7------5778
2026.05
33.157.224.441.628.426.420.4--