MongoDB, Inc. announced the integration of search and vector search capabilities with MongoDB Community Edition and MongoDB Enterprise Server. This announcement was made at its developer conference MongoDB.local NYC.
Previously exclusive to the fully managed MongoDB Atlas cloud platform, developers and organizations of all sizes can now access the preview of robust full-text search and vector search capabilities on MongoDB’s local, on-premises, and self-managed offerings. Starting today, these capabilities are in public preview for development and testing purposes.
“MongoDB sits at the gateway of meaning of an AI system,” said Dev Ittycheria, president and CEO, MongoDB. “Our platform is battle tested [for this].”
These native out-of-the-box search and AI-driven capabilities include full-text, semantic retrieval, and hybrid search to deliver highly accurate, intelligent, and context-aware retrieval-augmented generation (RAG) and agentic AI user experiences.
Previously, integrating search capabilities into self-managed MongoDB environments required adding on external search engines or vector databases. Managing a fragmented search stack added complexity and risk, and created operational overhead that could lead to fragile extract, transform, and load (ETL) pipelines, synchronization errors, and higher costs, according to the company.
Now, with search and retrieval capabilities directly integrated into MongoDB Community Edition and MongoDB Enterprise Server, developers and organizations can:
- Test and build AI applications locally: Vector search enables semantic information retrieval based on meaning encoded in vector embeddings. This empowers users to manage and build dynamic AI applications that rely on unstructured data such as text documents, images, videos, audio files, chat messages, and more, all within their local or on-premises environments.
- Boost accuracy with hybrid search: Combine keyword and vector search to return unified results from a single query for more accurate results. Crucial for reliable agentic solutions and AI applications, developers can easily take advantage of this powerful capability directly through MongoDB’s familiar query framework.
- Power AI agents with long-term memory: Allow data in MongoDB to serve as the long-term memory store for AI agents, enabling precise, context-aware applications ready for real-world situations. With Community Edition, developers can easily prototype RAG systems. Organizations building on Enterprise Server can securely ground AI agents in proprietary data on their own infrastructure.
Several MongoDB partners—including LangChain, a provider of software development frameworks for building LLM-powered applications, and LLamaIndex, an open-source framework for LLM applications—collaborated closely with MongoDB to test search and vector search capabilities in Community Edition.
MongoDB Search and MongoDB Vector Search are available in MongoDB Community Edition and Enterprise Server via public preview today.
Additionally, MongoDB introduced MongoDB 8.2, adding, what the vendor said, are significant performance improvements, including up to 49% faster unindexed queries and nearly triple the throughput for time-series bulk inserts.
The update also adds substring support to MongoDB’s queryable encryption, a client-side technology that encrypts sensitive data before it reaches the database and allows for expressive queries to be run on the encrypted data without the need for decryption.
Support for releases moving forward will also expand from 3 to 5 years.
For more information about this news, visit www.mongodb.com.