Introducing the Embedding and Reranking API on MongoDB Atlas

The next advancement in artificial intelligence (AI) involves enhancing the context in which AI models operate, emphasizing the importance of accurate data retrieval. As large language models (LLMs) become integral to various applications, the need for efficient search and retrieval systems becomes crucial. To address the complexity of building AI retrieval systems, MongoDB Atlas introduces the Embedding and Reranking API in collaboration with Voyage AI, offering developers access to advanced retrieval models. This API allows the creation of complete retrieval pipelines on a unified platform, supporting tasks from data storage to vector search, and embedding and reranking, with a flexible, token-based pricing model. Voyage AI's models, now integrated with MongoDB Atlas, are designed to enhance retrieval accuracy while optimizing for specific industry needs. The newly launched Voyage 4 model series introduces an innovative shared embedding space, providing developers with greater flexibility. MongoDB Atlas, trusted by a large number of enterprises, including over 75% of Fortune 100 companies, provides a secure, scalable platform for deploying AI applications, now augmented by the Embedding and Reranking API, enabling comprehensive AI retrieval capabilities.