MME

AI benchmark results and model performance

Type: percentageMultimodal

20.97%

19.15% - 22.53%

19.15% - DeepSeek VL2 Tiny

Scores Over Time

Individual benchmark scores plotted by date.

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
DeepSeek	DeepSeek VL2 Tiny	13 Dec 2024	19.15%	-	Yes	Source
DeepSeek	DeepSeek VL2 Small	13 Dec 2024	21.23%	-	Yes	Source
DeepSeek	DeepSeek VL2	13 Dec 2024	22.53%	-	Yes	Source