Multimodal Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
92.53Accuracy
73
May 28, 2026
84.6Accuracy
61
Jul 3, 2026
97.8Accuracy
61
Jun 1, 2026
70.5Accuracy
36
Feb 26, 2026
85.7MMBen
33
Jun 26, 2026
50.8F1 Score
32
Jun 11, 2026
86.3Accuracy
30
Jul 3, 2026
42.95B-1 Score
28
Apr 28, 2026
49.68Accuracy (Easy)
28
Apr 10, 2026
92.5Accuracy (Text)
27
Apr 23, 2026
89Accuracy
26
Apr 7, 2026
70.5Pass@1
24
May 7, 2026
68.3Total Score
24
Feb 26, 2026
55.93Average Score
22
Apr 7, 2026
90.92F1-Recall
22
Apr 7, 2026
52.6F1-Recall
22
Apr 7, 2026
39.98F1-Recall
22
Apr 7, 2026
55.47F1-Recall
22
Apr 7, 2026
53.9F1 Recall
22
Apr 7, 2026
0.902NAT Score
21
Feb 26, 2026
71.1Accuracy (All)
21
Feb 26, 2026
85.1MMBCN Score
20
Jun 26, 2026
49.31BLEU-1
20
May 1, 2026
57.4ROrig@k
20
May 28, 2026
57.7Perception (AD) Acc
19
Mar 10, 2026