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

MMStar

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal UnderstandingMMStar
Accuracy82
324
Multimodal ReasoningMMStar
Accuracy82
143
Multimodal EvaluationMMStar
Accuracy69.5
70
Visual Question AnsweringMMStar
Accuracy82.96
63
Image UnderstandingMMStar
Score65.1
54
General Visual Question AnsweringMMStar
Score77.8
35
General ReasoningMMStar
Score69.2
32
Multimodal UnderstandingMMStar
Average Score68.01
31
PerceptionMMStar latest (test)
CP67.2
30
General Visual ReasoningMMStar
Accuracy77.5
29
Multimodal ReasoningMMStar
Accuracy75.2
29
Multi-modal Visual CapabilityMMStar
Score63.9
29
Visual ReasoningMMStar
Accuracy68.2
27
General image understandingMMStar
Accuracy62.33
23
LVLM EvaluationMMStar
CP Score76.6
20
Visual PerceptionMMStar
Accuracy65.7
20
Multimodal Reasoning and PerceptionMMStar (test)
Accuracy63.9
19
Mathematical ReasoningMMStar Math
Accuracy77.2
19
Compositional ReasoningMMStar
Accuracy64.7
16
Multimodal ReasoningMMstar
Pass@1 Accuracy67.1
16
Vision-Language Perception and ReasoningMMStar
Accuracy (MMStar)39.9
16
Multimodal UnderstandingMMStar (test)
Score63.9
16
Visual UnderstandingMMStar
Accuracy (Clean)65.9
16
Multimodal PerceptionMMStar
Accuracy83.6
16
Multimodal IntegrationMMStar
Accuracy66.49
15
Showing 25 of 66 rows