MongoDB has announced the integration of search and vector search capabilities with its Community Edition and Enterprise Server, expanding access beyond its fully managed cloud platform, MongoDB Atlas.
This development enables developers and organisations of all sizes to access robust full-text search and vector search features for development and testing on self-managed deployments, including local and on-premises environments. These capabilities are now available in public preview.
Industry requirements
Modern application users increasingly expect personalised, high-performing, and real-time experiences. Developers therefore need comprehensive AI search and retrieval tools that operate natively within the databases storing their data.
The integration offers full-text, semantic retrieval, and hybrid search functionalities that support highly accurate and intelligent retrieval-augmented generation (RAG) and agentic AI user experiences. The update allows developers to test and build AI applications locally, make use of hybrid search to improve result accuracy, and equip AI agents with long-term memory through database integration.
"According to a 2025 IDC survey, more than 74% of organisations plan to use integrated vector databases to store and query vector embeddings within their agentic AI workflows," said Devin Pratt, Research Director at IDC. "In a fast-moving technological era driven by LLMs and AI applications, developers can't afford to be slowed down by fragmented systems. Embedding search and vector search directly into the database gives them one less complexity to manage, and allows them to stay focused on building intelligent applications."
Previously, users looking to include search and vector search within MongoDB's self-managed environments often relied on external engines or databases. This led to complexity and operational overhead, which risked issues such as fragile ETL (extract, transform, load) processes, synchronisation errors, and higher costs. The update aims to reduce these burdens by delivering native functionality.
Capabilities explained
The search and vector search integration allows several core workflows:
- Testing and building AI applications within local or on-premises environments, without external dependencies.
- Improved accuracy in search results by combining keyword-based and vector-based searches in a single query, vital for reliable agentic solutions.
- Powering AI agents with access to long-term memory, making use of MongoDB as a store for contextual or historical data that supports real-world applications.
Company perspective
"At MongoDB, we believe in empowering developers everywhere with the tools they need to build next-gen applications," said Ben Cefalo, Senior Vice President, Head of Core Products at MongoDB. "By expanding our Search and Vector Search capabilities, we're giving developers unparalleled flexibility to build in the environment of their choice, with the ultimate customer guarantee - that the core database and query capabilities they love in MongoDB Atlas are also freely available in Community. And when they're ready to bring their applications to market, they can easily migrate to our fully managed MongoDB Atlas platform for seamless scaling, multi-cloud flexibility, and enterprise-grade security."
MongoDB reports that the expanded feature set builds on its commitment to providing a unified document database platform capable of supporting operational data, search, real-time analytics, and AI-driven data retrieval in a single environment.
Partner validation
Several MongoDB partners participated in early testing of the search and vector search features in Community Edition. These include LangChain and LlamaIndex, both of which focus on application development for large language models (LLMs).
"We're thrilled MongoDB search and vector search are now accessible in the already popular MongoDB Community Edition," said Harrison Chase, CEO, LangChain. "Now our customers can leverage MongoDB and LangChain in either deployment mode and in their preferred environment to build cutting edge LLM applications."
"We're excited about the next interaction of search experiences in MongoDB Community Edition. Our customers want the highest flexibility to be able to run their search and gen AI-enabled applications, and bringing this functionality to Community unlocks a whole new way to build and test anywhere," said Jerry Liu, CEO, LlamaIndex.
The new features are available in public preview via MongoDB Community Edition and MongoDB Enterprise Server. The company aims to support developers building intelligent AI applications that provide relevant context for agentic systems in their environment of choice.
