Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended image description on GPT-4o assisted evaluation
Loading...
8.76
Accuracy
R-CoV
4.6208
5.6954
6.77
7.8446
Apr 22, 2026
Accuracy
Relevance
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Relevance
R-CoV
Model=Qwen2.5-VL
2026.04
8.76
9.72
Vanilla
Model=Qwen2.5-VL
2026.04
8.35
9.65
R-CoV
Model=LLaVA-1.5
2026.04
7.48
9.03
R-CoV
Model=MiniGPT-4
2026.04
7.24
8.9
R-CoV
Model=mPLUG-Owl
2026.04
6.91
8.41
Vanilla
Model=LLaVA-1.5
2026.04
6.48
8.91
Vanilla
Model=MiniGPT-4
2026.04
5.69
8.62
Vanilla
Model=mPLUG-Owl
2026.04
4.78
7.94
Feedback
Search any
task
Search any
task
