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

Vision-Language Capability Evaluation on MME

2,349Score

Vanilla

-34.472584.3141,203.11,821.886Feb 10, 2026Feb 21, 2026Mar 5, 2026Mar 16, 2026Mar 28, 2026Apr 8, 2026Apr 20, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
2,349
2026.04
2,287
2026.04
2,252
2026.04
2,251
2026.04
2,189
2026.02
1,594
2026.02
1,588
2026.02
1,569
2026.02
1,564
2026.02
1,511
2026.02
1,477
2026.02
1,466
2026.02
1,440
2026.03
1,298
2026.03
1,296
2026.03
1,263
2026.04
76.3
2026.04
74.1
2026.04
73.4
2026.04
70.9
2026.03
63.8
2026.03
62.4
2026.03
62.1
2026.03
61.9
2026.03
61.9
2026.03
60.6
2026.03
60.5
2026.03
60.1
2026.03
59.5
2026.03
59.4
2026.03
57.2