Multimodal Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.118UPR
6
Feb 26, 2026
56.7Accuracy
5
Jun 5, 2026
32Accuracy
5
Jun 5, 2026
54.7RC Accuracy Avg
5
Apr 23, 2026
39RC-Acc
5
Apr 23, 2026
57.8RC-Acc
5
Apr 23, 2026
74.8RC-Acc
5
Apr 23, 2026
14.5EG Score
5
Feb 26, 2026
62.53Accuracy
4
Jul 3, 2026
44.2Accuracy (SEED)
4
Jun 12, 2026
52.32Score (%)
4
Jun 9, 2026
34.05MMMU-P Score
4
May 28, 2026
3.985MM Intent Score
4
May 20, 2026
0.364Deductive Score
4
Feb 26, 2026
64Average Sample Size (N)
3
Jul 3, 2026
64Average N
3
Jul 3, 2026
64Average N
3
Jul 3, 2026
37.23Accuracy
3
Jun 23, 2026
29.8Accuracy
3
Jun 16, 2026
20.5Accuracy
3
Jun 16, 2026
31.3Accuracy
3
Jun 16, 2026
11MathVersion
3
Apr 23, 2026
29Accuracy (MathVision)
3
Apr 10, 2026
69Pass@1 Accuracy
3
Feb 26, 2026
46Pass@1 Acc
3
Feb 26, 2026