Jina CLIP v2: Multilingual & Multimodal Search

**What the feature is**
jina-clip-v2 is a general-purpose multilingual multimodal embedding model for text & images designed for building advanced search and retrieval systems.

**Value proposition**
By jointly encoding both text and images into a shared vector space, jina-clip-v2 enables powerful cross-modal applications such as searching images using text queries (and vice versa) across 89 supported languages. This version is a significant upgrade over its predecessor, featuring a 512x512 input resolution for capturing finer visual details and supporting Matryoshka Representations, which allow you to truncate embedding dimensions (from 1024 down to 64) to significantly reduce storage costs and latency while maintaining high accuracy.