Overview
This node inserts vector data into a Zilliz vector database collection. It is designed to facilitate adding new vectors along with optional textual content and metadata fields into a specified collection within a Zilliz database.
Common scenarios where this node is beneficial include:
- Enriching a vector search index by adding new embeddings generated from text or other data.
- Maintaining an up-to-date vector store for similarity search applications such as recommendation systems, semantic search, or machine learning pipelines.
- Clearing existing data in a collection before bulk inserting fresh vectors to reset the dataset.
Practical example:
- You have processed documents and generated vector embeddings for each document. Using this node, you can insert those vectors into a Zilliz collection named "my_collection" in the "default" database, optionally including the original text and additional metadata fields like author or timestamp.
Properties
| Name | Meaning |
|---|---|
| Database Name | Name of the Zilliz database to use (default is "default"). |
| Collection Name | Name of the Zilliz collection to insert vectors into. Must start with a letter and contain only letters, numbers, and underscores. |
| Options | A set of additional options: |
| - Clear Collection | Boolean flag indicating whether to clear all existing vectors in the collection before inserting new data. |
| - Text Field | Name of the field in input data containing the text content to be stored alongside the vector (default "text"). |
| - Vector Field | Name of the field in input data containing the vector data to insert (default "vector"). |
| - Metadata Fields | Comma-separated list of additional fields from input data to include as metadata in the inserted vector record. |
Output
The node outputs an array of JSON objects, one per input item, each containing:
success: Boolean indicating if the insertion succeeded.insertCount: Number of vectors successfully inserted (usually 1 per item).insertIds: Array of IDs assigned to the newly inserted vectors.collection: The name of the collection where vectors were inserted.database: The database name used.
If an error occurs during processing an item and the node is configured to continue on failure, the output for that item will contain an error field with the error message.
The node does not output binary data.
Dependencies
- Requires an API key credential and cluster endpoint for authenticating with the Zilliz vector database service.
- Uses an internal Zilliz client library to communicate with the vector database.
- The user must configure the node with valid credentials providing access to the Zilliz service.
Troubleshooting
- Vector field not found: If the specified vector field is missing in the input data, the node throws an error indicating the missing field. Ensure the input items contain the vector data under the correct field name.
- Collection clearing errors: If clearing the collection fails (e.g., due to permissions), the node ignores the error silently but may affect subsequent insertions. Verify that the API key has sufficient rights to delete vectors.
- Invalid collection name: Collection names must start with a letter and contain only letters, numbers, and underscores. Invalid names will cause validation errors.
- Credential issues: Missing or invalid API keys or endpoints will prevent connection to the Zilliz service. Confirm credentials are correctly set up in n8n.
- Continue on Fail: When enabled, the node continues processing remaining items even if some fail, returning error details per failed item.
Links and References
- Zilliz Vector Database Documentation (for general understanding of the service)
- n8n Documentation on Credentials (for setting up API keys)
- Vector Search Concepts (background on vector databases and similarity search)