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

MMStar

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal UnderstandingMMStar
Accuracy82
511
Multimodal EvaluationMMStar
Accuracy69.5
177
Visual Question AnsweringMMStar
Accuracy91.9
151
Multimodal ReasoningMMStar
Accuracy82
143
Multimodal ReasoningMMStar
Accuracy79.2
102
General image understandingMMStar
Accuracy72.13
67
Image UnderstandingMMStar
Score65.1
56
General Visual ReasoningMMStar
Accuracy77.5
55
General VQAMMStar
Accuracy80.1
51
Visual ReasoningMMStar
Accuracy69
51
General Visual Question AnsweringMMStar
Score77.8
50
General TaskMMStar
Accuracy76.2
36
Visual Question AnsweringMMStar (test)
Accuracy68
34
General ReasoningMMStar
Score69.2
32
Multimodal Capability EvaluationMMStar
Overall Score68.3
31
Multimodal UnderstandingMMStar
Average Score68.01
31
Visual PerceptionMMStar
Accuracy73.07
30
PerceptionMMStar latest (test)
CP67.2
30
Multimodal ReasoningMMStar
Accuracy75.2
29
Multi-modal Visual CapabilityMMStar
Score63.9
29
Multi-modal ReasoningMMStar
Accuracy63.78
28
Image ReasoningMMStar
Accuracy71.85
27
Multimodal UnderstandingMMStar
Score68.33
26
Mathematical ReasoningMMStar Math
Accuracy77.2
26
Multimodal UnderstandingMMStar (test)
Accuracy71.6
26
Showing 25 of 94 rows