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

Qwen2-VL

Benchmarks

Task NameDataset NameSOTA ResultTrend
Overall Multimodal EvaluationQwen2-VL 7B Evaluation Suite
Relative Accuracy100
5
Keyword Matching AttackQwen2-VL-7B
KMR Alpha89.9
4
Energy consumption rankingQwen2-VL image workload
Pairwise Accuracy89.3
2
Energy consumption rankingQwen2-VL text workload
Pairwise Accuracy95.8
2
Showing 4 of 4 rows