Sigmoid Loss for Language Image Pre-Training
Paper
•
2303.15343
•
Published
•
8
SigLIP release
SigLIP improves upon CLIP with a sigmoid loss. Both English-only and multilingual checkpoints are released.
-
-
google/siglip-base-patch16-224
Zero-Shot Image Classification • Updated • 187k • 50 -
google/siglip-base-patch16-256
Zero-Shot Image Classification • Updated • 11k • 6 -
google/siglip-base-patch16-384
Zero-Shot Image Classification • Updated • 3.12k • 10 -
google/siglip-base-patch16-512
Zero-Shot Image Classification • Updated • 126k • 24 -
google/siglip-large-patch16-256
Zero-Shot Image Classification • Updated • 20.2k • 12 -
google/siglip-large-patch16-384
Zero-Shot Image Classification • Updated • 17.8k • 9
google/siglip-base-patch16-256-multilingual
Zero-Shot Image Classification • Updated • 20.1k • 44Note This is a multilingual variant trained on WebLi without language filtering.
google/siglip-so400m-patch14-384
Zero-Shot Image Classification • Updated • 5.86M • 547Note This one has a scaled-optimized architecture, obtaining the best performance in English.