Multimodal model · Established · Advanced

CLIP

A contrastively trained model that places images and text in a shared embedding space, enabling zero-shot classification and cross-modal search.

What CLIP is

CLIP learns from image-caption pairs to score how well a picture matches a description, which allows classification against arbitrary labels with no task-specific training.

How it works

Separate image and text encoders are trained so matching pairs align and mismatched pairs separate. The resulting embeddings power search, filtering and conditioning for generative models.

Why it matters

It is the connective component behind text-to-image conditioning and most image search systems built after 2021.

Common uses

→Image search by description
→Zero-shot image classification
→Dataset filtering
→Guidance for image generation

Strengths

✓Zero-shot flexibility
✓Reusable embeddings

Watch for

✓Inherits caption-data biases
✓Weak on fine-grained detail and text in images

Continue exploring

Contrastive Learning Embeddings Multimodal AI AI Image Generation