Developers utilizing Elastic for search and retrieval applications can now access Jina AI's latest embedding and reranking models through the Elasticsearch Open Inference API. This integration enables the use of high-performance tools for information retrieval, enhancing both semantic search and RAG applications.

Jina AI's models include support for multilingual text embeddings and reranking, optimizing them for various applications including retrieval, clustering, and classification. Dr. Saahil Ognawala, head of product at Jina AI, stated that the integration delivers enterprise-grade semantic search to production settings and supports open-source communities.

Ajay Nair, general manager at Elastic, noted the collaboration allows users to leverage Jina’s efficient tools alongside Elastic’s existing models, streamlining the process for developers aiming to create generative AI applications. Support for these Jina AI models is available immediately on Elastic Cloud Serverless.

Developers can access various supported models, including jina-embeddings-v3 and jina-reranker-v2-base-multilingual. For a comprehensive list of supported models, additional information is available in the Elastic blog.