MME
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MME
86.3Accuracy@1
18
MME
1,510Perception Score (P)
18
MME (total)
1,699MME-P Score
18
MME
100Average Score
18
MME Perception 16
1,515.3Perception Score
18
MME Existence (test)
96.67Accuracy
18
MME Leaderboard (full)
138.33Existence Score
18
MME-RW
68.6Accuracy
17
MME RW-L
12.2C@5
17
MME
1,842MME Score
17
MME RWLite
52.6Score
17
MME (test)
136.43Communication Score
17
MME-RealWorld
51.94Score
17
MME pt-PT
2,355Score
16
MME
1,696.1MME Score
16
MME
2,388MME Score
16
MME
2,433.9Score
16
MME hallucination subset
195Existence Score
16
MME
189.47Existence Score
16
MME
200Existence Score
16
MME full
2,418.5MME Score
16
MME (test)
190Existence
15
MME Perception
1,638MME-P Score
15
MME Real-World
38.9Score
15
MME v1 (full)
1,657.9Score
15