Multimodal Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
86.2MM-Vet Score
551
Jul 3, 2026
83.89Accuracy
220
Jul 3, 2026
78Accuracy
199
Jun 30, 2026
90.63Accuracy
180
Jun 30, 2026
84.78Accuracy
172
Jul 3, 2026
85.6Accuracy
171
Jul 1, 2026
78.2Accuracy
168
May 11, 2026
59.41Accuracy
162
May 27, 2026
82Accuracy
143
Mar 31, 2026
68.5Accuracy
138
Jun 23, 2026
80.68Accuracy
125
Jun 2, 2026
82Accuracy
119
Jun 24, 2026
79.2Accuracy
102
Jul 7, 2026
85.9Accuracy
89
Jun 26, 2026
72.9Accuracy
77
Jul 1, 2026
67.2Accuracy
77
Jun 23, 2026
58.7Accuracy
70
Mar 24, 2026
84.29MMBench Accuracy (en)
65
Jul 1, 2026
78.6Score
60
Jun 2, 2026
78.1Accuracy
59
Jul 7, 2026
91.61Total Acc
55
May 28, 2026
83.5Accuracy
50
Jul 3, 2026
87.6Accuracy
47
Feb 26, 2026
97.85Average Accuracy
45
May 27, 2026
64.7Accuracy
43
Jun 30, 2026