n8n and Zilliz Cloud Integration
Integrate Zilliz Cloud with n8n to build AI-powered workflows with semantic search, RAG, and agents using a visual no-code automation builder.
Use this integration for FreeWhat is n8n
n8n is a powerful open-source workflow automation platform that allows you to connect various applications, services, and APIs together to create automated workflows without coding. With its node-based visual interface, n8n enables users to build complex automation processes by simply connecting nodes that represent different services or actions. It is self-hostable, highly extensible, and supports both fair-code and enterprise licensing.
By integrating Zilliz Cloud(fully managed Milvus) with n8n, you can bring the power of high-performance vector search and AI into your automation workflows.
Benefits of the n8n + Zilliz Cloud Integration
Integrating Zilliz Cloud (managed Milvus) with n8n removes the complexity of building AI-powered workflows from scratch. You can set up semantic search, RAG pipelines, and intelligent agents in minutes—without writing any code. With Zilliz Cloud handling the vector database infrastructure, you can focus on building, not maintaining. And because everything runs inside n8n's visual workflow editor, your entire AI stack stays in one place, easy to manage and scale.
How the Integration Works
The Milvus Vector Store node acts as the bridge between n8n and Zilliz Cloud. When a workflow runs, the node connects to your Zilliz Cloud instance and handles all vector operations—inserting documents, querying by semantic similarity, and returning relevant results—directly within the workflow pipeline.
Depending on your use case, the node can be configured in four operation modes: Get Many, Insert Documents, Retrieve Documents (for chains), and Retrieve Documents (for AI agents). This makes it flexible enough to serve as a standalone data store, a retriever connected to a question-answering chain, or a live knowledge tool plugged directly into an AI agent.
Additional options like metadata filtering and collection management give you fine-grained control over how data is stored and retrieved in Zilliz Cloud.
Step-by-Step: Setting Up the Milvus Vector Store Node in n8n
The Milvus Vector Store node supports four usage patterns. Choose the one that fits your workflow.
Pattern 1: Insert and Retrieve Documents (Regular Node)
Use the Milvus Vector Store as a standalone node in a regular workflow—no AI agent involved.
- Set the operation mode to Insert Documents to store content into a Milvus collection.
- Set the operation mode to Get Many to retrieve documents based on semantic similarity to a query prompt.
- Optionally, enable Clear Collection before insertion to remove existing data first.
- Use Metadata Filter in Get Many mode to narrow results by custom metadata fields (multiple filters apply AND logic).
This pattern is ideal for building document pipelines that store and retrieve content for cited, chat-based answers.
Pattern 2: Connect Directly to an AI Agent as a Tool
Plug the Milvus Vector Store directly into an AI agent's tool connector so the agent can query the vector store autonomously when answering questions.
Connection flow:
AI Agent (tools connector) → Milvus Vector Store
- In the Milvus Vector Store node, set the operation mode to Retrieve Documents (As Tool for AI Agent).
- Connect the node to the tools connector of your AI Agent node.
- The agent will decide when to query Milvus based on the user's question, using it as a dynamic knowledge resource.
This pattern works well when you want the AI agent to have flexible, on-demand access to a knowledge base.
Pattern 3: Use a Retriever with a Q&A Chain
Pair the Milvus Vector Store with a Vector Store Retriever and a Question and Answer Chain to build a structured Q&A system.
Connection flow:
Question and Answer Chain (Retriever connector) → Vector Store Retriever (Vector Store connector) → Milvus Vector Store
- Set the Milvus Vector Store operation mode to Retrieve Documents (As Vector Store for Chain/Tool).
- Connect it to the Vector Store connector of the Vector Store Retriever node.
- Connect the Retriever to the Retriever connector of the Question and Answer Chain node.
- The chain will automatically fetch the most relevant documents from Milvus to answer the user's input.
This pattern is best for structured document Q&A where you want precise retrieval before generating an answer.
Pattern 4: Use the Vector Store Question Answer Tool
Instead of connecting Milvus directly as a tool, wrap it with a Vector Store Question Answer Tool node. This tool summarizes the retrieved content before passing it to the agent.
Connection flow:
AI Agent (tools connector) → Vector Store Question Answer Tool (Vector Store connector) → Milvus Vector Store
- Set the Milvus Vector Store operation mode to Retrieve Documents (As Vector Store for Chain/Tool).
- Connect it to the Vector Store connector of the Vector Store Question Answer Tool node.
- Connect the QA Tool to the tools connector of the AI Agent.
- The tool handles retrieval and summarization automatically, giving the agent a cleaner, condensed answer to work with.
This pattern is useful when you want the agent to receive a summarized answer from the knowledge base rather than raw document chunks.
Learn More
- Milvus × n8n Integration Guide
- Build a document QA system with RAG using Milvus, Cohere, and OpenAI for Google Drive
- Milvus node settings - Feature Requests
- n8n Milvus Integration Documentation
- I Discovered This N8N Repo That Actually 10x'd My Workflow Automation Efficiency
- https://zilliz.com/product/integrations