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

MMK12

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal Math ReasoningMMK12
Accuracy80.83
41
Multimodal ReasoningMMK12
Accuracy81.8
29
Multi-modal ReasoningMMK12 (test)
Accuracy73.9
28
Multimodal ReasoningMMK12
Pass@151
17
Visual Mathematical ReasoningMMK12 (val)
Accuracy58.86
9
Scientific ReasoningMMK12 Chem
Accuracy63.7
8
Mathematical ReasoningMMK12 Math
Accuracy64.7
8
Mathematical and Scientific ReasoningMMK12
MMK12 Math Score65.1
8
Multidisciplinary ReasoningMMK12
Accuracy75.2
6
Showing 9 of 9 rows