MME
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MME Cognition
603.57MME-c Score
10
MME v1.0 (test)
100Relative Performance
10
MME 80% (test)
87.89Accuracy
10
MME
3.78Avg Accepted Length (tau)
10
MME-RW
31.9Mean Accuracy
10
MME RealWorld
65.3Score
10
MME v1.0 (test)
1,531.3MME^P Score
10
MME Cognition 16
399.5Cognition Score
10
MME-R
49.71MME-R Score
9
MME-RealWorld in-distribution (ID) and held-out categories
84.38Diagram Accuracy
9
MME
2,325MME-T Score
9
MME-r reasoning
722.5Score
9
MME-RealWorld
65.8Accuracy
9
MME
90.85M(cl, nc)
9
MME-Unify
13Spot Difference
9
MME
78.7Normalized Score
8
MME (test)
2,448Sum Score
8
MME
655.7Cognition Score
8
MME
-53.47BD-Rate
8
MME
1,950MME Score
8
MME (full)
2,457.1Overall Score
8
MME (test)
2,473.6Score
8
MME
1,851Total Score
8
MME-C
561Cognition Score
8
MME
530.7Cognition Score
8