AI engineering glossary · Generative media multimodal
What is sigLIP?
SigLIP: A CLIP-style image-text model that trains with a pairwise sigmoid loss, treating every image-text pair as an independent yes/no decision instead of using a softmax over the whole batch. It scales more easily and is a common vision encoder in vision-language models.
SigLIP: an example
A vision-language model uses a SigLIP encoder to turn each input image into patch features for the language model.
Related terms
More Generative media multimodal terms
- Speaker diarisation
- Speaker embedding
- Spectrogram
- Speech-to-speech translation
- SSIM
- Stable Diffusion
- Streaming ASR
- StyleGAN
Where you learn it
SigLIP is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.