Multimodal Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
58.75Accuracy
12
Jun 5, 2026
69.51Accuracy
12
May 5, 2026
37.3Accuracy
12
May 5, 2026
36.67Accuracy
12
May 5, 2026
68.33Accuracy
12
May 5, 2026
55.22BoN@8 Accuracy
12
Mar 18, 2026
88.5Average Score
12
Feb 26, 2026
90.6Accuracy
12
Feb 27, 2026
60.4Score
12
Feb 27, 2026
10.9Contradiction Rate
12
Feb 26, 2026
11.3Contradiction Rate
12
Feb 26, 2026
5Contradiction Rate
12
Feb 26, 2026
80.8Average Score
11
Jul 2, 2026
55.33Accuracy
11
Jun 16, 2026
29.4MathVision Score
11
May 18, 2026
63.7MathVista Score
11
May 18, 2026
54.67Strict Accuracy
11
Apr 21, 2026
73.6Strict Accuracy
11
Apr 21, 2026
67.1Accuracy (ACC%)
11
Mar 17, 2026
32.52ECE
11
Feb 26, 2026
35.4Aerial Score
11
Feb 26, 2026
49.2Mean@1
11
Feb 26, 2026
6.67Mean@1
11
Feb 26, 2026
48.23BLEU
11
Feb 26, 2026
59.23Accuracy
11
Feb 26, 2026