Multimodal Models — CLIP & BLIP

Models that bridge vision and language: CLIP, BLIP, and multimodal embeddings for search and generation.