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

Comprehensive Multimodal Evaluation on SEED-Bench Image

77.3Accuracy

SAIL-VL2

63.2666.90570.5574.195Dec 16, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
77.3
2025.12
75.3
2025.12
74.8
2025.12
74.6
2025.12
74.2
2025.12
74.1
2025.12
73.5
2025.12
73.3
2025.12
71.6
2025.12
63.8