Jina AI

Jina Embeddings are a family of high-performance, open-source, multilingual, and multimodal embedding models developed by Jina AI. They transform text and visual data (like images and charts) into dense numerical vectors for semantic search, RAG, and AI applications. Key features include supporting long contexts (up to 32k tokens), multi-task capabilities, and specialized versions for text-to-image and code retrieval.

Visit the following resources to learn more:

• @official@Jina Embeddings
• @official@https://jina.ai/news/jina-embeddings-v5-text-distilling-4b-quality-into-sub-1b-multilingual-embeddings/