Skip to content

SigLIP 400M

Google DeepMindImage classificationImage embeddingOpen weights

SigLIP 400M is image classification model published by Google DeepMind in 2023 featuring 400000000.0 parameters.

About SigLIP 400M

We propose a simple pairwise Sigmoid loss for Language-Image Pre-training (SigLIP). Unlike standard contrastive learning with softmax normalization, the sigmoid loss operates solely on image-text pairs and does not require a global view of the pairwi

Details

Provider
Google DeepMind
Task
Image classification,Image embedding
Parameters
400000000.0
Released
2023-03-27
Open weights
Yes
View model source

Explore

FAQ