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

Multimodal Understanding on DeepSeek-VL2 Evaluation Suite

72.68Average Score

Baseline

66.367268.006169.64571.2839Nov 22, 2025
Updated 26d ago

Evaluation Results

MethodLinks
2025.11
72.6800
2025.11
72.490.1921.09
2025.11
72.180.521.09
2025.11
71.940.7443.07
2025.11
71.910.7743.07
2025.11
71.720.9621.09
2025.11
70.322.3663.66
2025.11
70.312.3743.07
2025.11
70.022.6663.66
2025.11
66.616.0763.66