MongoDB and partners have developed specific product integrations to help you leverage MongoDB in your AI-powered applications and AI agents.
This page highlights AI integrations that MongoDB and partners have developed. You can connect MongoDB to popular AI providers and LLMs through their standard connection methods and APIs. For a complete list of integrations and partner services, see the MongoDB Partner Ecosystem Catalog.
This page also lists community-maintained integrations. These integrations are developed on an open-source basis and not officially managed by MongoDB. Community-maintained integrations are labeled accordingly in the description column. For support with these integrations, contact the given project's maintainers.
AI Application and Agent Frameworks
The following frameworks use MongoDB as a vector store, memory backend, or feature store to support AI applications and agents. Use cases include RAG, agentic memory, multi-agent workflows, and machine learning services.
Framework | Description | Documentation |
|---|---|---|
Python framework for building autonomous AI agents with specialized roles and multi-agent applications with "crews" that can complete complex tasks by delegating work amongst themselves. | ||
Feature store framework that allows you to use MongoDB as both an online and offline store for defining, storing, and serving ML features. | ||
Python framework for building custom applications with LLMs, embedding models, vector search, and more for use cases such as RAG. | ||
Framework for building AI applications by using "chains," LangChain-specific components that can be combined together for various use cases. The LangChain MongoDB integration provides several components for RAG. | ||
Brings LangChain capabilities to Java. | ||
Brings LangChain capabilities to the Go ecosystem. | ||
Specialized framework within the LangChain ecosystem for building AI agents and complex multi-agent workflows, with support for persistence, streaming, and memory. | ||
Framework that provides several tools for connecting custom data sources to LLMs and building RAG applications. | ||
Open-source PHP framework for building AI-powered applications. Supports MongoDB as a vector store with Voyage AI embedding models for use cases such as RAG. | ||
Open-source TypeScript framework that provides components for building AI agents, including workflows, RAG, and evals. Uses MongoDB for vector storage and retrieval, RAG, and memory. | ||
Framework from Microsoft that combines various AI services with your applications for use cases including RAG. | ||
Framework that applies Spring design principles to AI applications for use cases including RAG. | ||
COMMUNITY MAINTAINED - Fork of AutoGen for building multi-agent AI applications. Uses MongoDB as a vector store for embeddings and retrieval-augmented generation. | ||
COMMUNITY MAINTAINED — Framework for building collaborative AI agent systems. Uses MongoDB for semantic and episodic vector memory and session persistence. | ||
COMMUNITY MAINTAINED — Python framework for building multi-modal AI agents. Uses MongoDB for memory storage, session persistence, workflow state, and vector search. | ||
COMMUNITY MAINTAINED — Microsoft framework for building multi-agent AI applications. Uses MongoDB as a vector store for retrieval-augmented generation. | ||
COMMUNITY MAINTAINED — Open-source LLM application development platform. Uses bundled LangChain nodes to support MongoDB for vector storage and chat message history. | ||
COMMUNITY MAINTAINED — Python framework for building stateful AI applications and agents. Use MongoDB to persist agent state, sessions, and workflow checkpoints. | ||
COMMUNITY MAINTAINED — Library for representing and processing multi-modal data. Use MongoDB Atlas as a document index and vector store. | ||
COMMUNITY MAINTAINED — Open-source tool for querying documentation with AI. Uses MongoDB as a vector store, document store, and conversation history backend. | ||
COMMUNITY MAINTAINED — Framework for evolving and automating AI agents. Uses MongoDB as a tool registry, artifact store, and workflow state database. | ||
COMMUNITY MAINTAINED — Framework for graph-based retrieval-augmented generation. Use MongoDB for document storage, index storage, and vector retrieval. | ||
COMMUNITY MAINTAINED — Framework for building enterprise-grade RAG and agent pipelines. Use MongoDB Atlas as an embedding vector store. | ||
COMMUNITY MAINTAINED — Memory layer for AI agents and assistants. Uses MongoDB Atlas as a vector store for persistent agent memory. | ||
COMMUNITY MAINTAINED — AI memory infrastructure platform. Uses MongoDB as a bring-your-own-database backend for memory and session storage. | ||
COMMUNITY MAINTAINED — Microsoft framework for building enterprise AI agents. Uses MongoDB as a vector store in C# and Python agent applications. | ||
COMMUNITY MAINTAINED — Agent framework for building backend AI resources. Uses MongoDB as a backend document and data store for your agents. | ||
COMMUNITY MAINTAINED — Official SDK from OpenAI for building multi-agent applications. Use MongoDB for persistent session and memory storage between agent turns. | ||
COMMUNITY MAINTAINED — Python agent framework built on Pydantic. Uses MongoDB Atlas-backed persistent memory for your AI agents. | ||
COMMUNITY MAINTAINED — Open-source framework for building conversational AI and chatbots. Use MongoDB as a tracker store for conversation state persistence. | ||
COMMUNITY MAINTAINED — Rust framework for building LLM-powered applications. Uses MongoDB as a document store and vector store index. | ||
COMMUNITY MAINTAINED — Open-source AI application builder. Uses MongoDB Atlas as a vector store component in your AI workflows. | ||
COMMUNITY MAINTAINED — Alibaba's extension of Spring AI for building AI applications. Use MongoDB as a checkpoint saver and workflow state store. | ||
COMMUNITY MAINTAINED — AWS-backed SDK for building AI agents. Uses MongoDB Atlas with Bedrock embeddings for vector-based memory storage and retrieval. | ||
COMMUNITY MAINTAINED — Python framework for building data and AI applications. Uses MongoDB as a data node for reading and writing collection data. | ||
COMMUNITY MAINTAINED — Python framework for building reliable AI agents. Uses MongoDB for session and workflow state storage across agent runs. |
Platforms and Automations
You can also integrate with the following enterprise platforms to build generative AI applications. These platforms offer features such as pre-trained models, workflow automation tools, visual pipeline builders, and backend-as-a-service platforms using MongoDB as a data layer.
Platform | Description | Documentation |
|---|---|---|
Fully-managed platform for building generative AI applications. Integrate MongoDB as a knowledge base to store custom data in MongoDB Atlas, implement RAG, and deploy agents. | ||
Platform from Google Cloud for building and deploying AI applications and agents. Includes tools and pre-trained models from Google that you can use with MongoDB Atlas for RAG and other use cases such as natural language querying. | ||
No-code workflow automation tool that enables you to build agentic workflows through interactive nodes in a visual canvas. Supports multiple MongoDB nodes, including nodes for RAG and memory for your AI agents. | ||
COMMUNITY MAINTAINED — Open-source backend-as-a-service platform. Uses MongoDB as its default database backend for document, session, and workflow state storage. | ||
COMMUNITY MAINTAINED — Platform for building and deploying AI tools and function-calling integrations. Uses the MongoDB toolkit to give agents access to your data. | ||
COMMUNITY MAINTAINED — Backend-as-a-service platform built on Parse Server. Uses MongoDB as the default database for app data, documents, and session management. | ||
COMMUNITY MAINTAINED — Open-source workflow automation platform. Uses MongoDB as a connector for CRUD operations, vector storage, and chat memory in workflows. | ||
COMMUNITY MAINTAINED — Open-source LLM application development platform. Uses MongoDB Atlas as a vector store, document store, and retrieval node in AI workflows. | ||
COMMUNITY MAINTAINED — Open-source low-code platform for building LLM applications. Use MongoDB Atlas for vector storage and chat memory in visual workflows. | ||
COMMUNITY MAINTAINED — Open-source platform for building AI knowledge bases and chat applications. Uses MongoDB for session and application state storage. | ||
COMMUNITY MAINTAINED — Open-source tool for curating and analyzing visual datasets. Use MongoDB Atlas as a vector similarity index for dataset search. | ||
COMMUNITY MAINTAINED — Open-source visual framework for building AI pipelines and agents. Uses MongoDB Atlas as a vector store component. | ||
COMMUNITY MAINTAINED — Open-source AI chat platform. Uses MongoDB as its default database for conversation history, messages, and user session management. | ||
COMMUNITY MAINTAINED — No-code workflow automation platform. Uses the native MongoDB app to integrate your MongoDB data in automated workflows. |
ETL and Data Pipelines
The following tools support extracting, transforming, and loading data with MongoDB as a source or destination connector for your AI and data pipelines.
Tool | Description | Documentation |
|---|---|---|
COMMUNITY MAINTAINED — Open-source data integration platform for ETL and ELT pipelines. Uses MongoDB as a source or destination connector. | ||
COMMUNITY MAINTAINED — Open-source platform for scheduling and monitoring data workflows. Use MongoDB to ingest or query data from your pipelines. | ||
COMMUNITY MAINTAINED — Open-source Python library for building data pipelines. Uses the MongoDB source to extract and load data into your destination of choice. | ||
COMMUNITY MAINTAINED — Enterprise integration platform for connecting applications and data. Uses the MongoDB Snap Pack for data ingestion, queries, and ETL workflows. | ||
COMMUNITY MAINTAINED — Data preprocessing platform for ingesting unstructured content into AI pipelines. Uses MongoDB as an ingestion destination connector. |
Tools
The following tools extend your AI applications with connectivity, observability, and memory capabilities. These include MCP servers for agent-to-data connectivity, LLM gateways and proxies, and observability platforms for tracing and evaluating LLM applications.
Tool | Description | Documentation |
|---|---|---|
Model Context Protocol (MCP) is an open standard for how LLMs connect to and interact with external resources and services. Uses MongoDB's official MCP Server implementation to interact with your MongoDB data and clusters from your agentic AI tools. | ||
COMMUNITY MAINTAINED — Open-source observability and evaluation platform for LLM applications. Uses MongoDB Atlas Vector Search as a traced vector database target. | ||
COMMUNITY MAINTAINED — AI gateway for managing LLM traffic and observability. Use MongoDB as a log and trace store for enterprise deployments. | ||
COMMUNITY MAINTAINED — Registry and hosting platform for MCP servers. Uses the official MongoDB MCP server from Smithery to connect agents to your data. | ||
COMMUNITY MAINTAINED — AI-optimized search API for agents and RAG. Uses MongoDB Atlas as the backend for hybrid search with the Tavily Hybrid Client. |