Actions11
Overview
This node provides advanced integration with the Qdrant vector search engine API, allowing users to manage collections and points within those collections. It supports a wide range of operations such as creating, updating, deleting collections and points, searching points by vector similarity, retrieving points by IDs, counting points, and listing collections.
Typical use cases include:
- Managing vector collections for machine learning or AI applications.
- Performing similarity searches on embedding vectors.
- Maintaining and updating large datasets of vectorized data points.
- Integrating Qdrant's vector search capabilities into automated workflows.
For example, you can use this node to create a new collection configured for your specific vector dimensions, upsert points representing documents or images with their embeddings, and then perform similarity searches to find related items.
Properties
| Name | Meaning |
|---|---|
| Operation | The action to perform on Qdrant. Options: Count Points, Create Collection, Delete Collection, Delete Points, Get Collection Info, Get Points by IDs, List Collections, Search Points, Update Collection, Update Points, Upsert Points. |
| Collection Name | The name of the Qdrant collection to operate on. Supports expressions. Required for most operations. |
| Collection Config (JSON) | Raw JSON configuration for creating or updating a collection (e.g., vector size, replication factor). Used only for Create Collection and Update Collection operations. Supports expressions. |
| Point IDs (JSON Array) | Array of point IDs (e.g., [1,2,3]) used for retrieving or deleting points. Used in Get Points and Delete Points operations. Supports expressions. |
| Points (JSON Array) | Array of point objects with structure { ID, vector: [...], payload: {...} } for inserting or updating points. Used in Upsert Points and Update Points operations. Supports expressions. |
| Search Vector (JSON Array) | Embedding vector used to perform similarity search against the collection. Required for Search Points operation. Supports expressions. |
| Filter (JSON) | Filter object to narrow down points during search or count operations. Supports expressions. |
| Limit | Maximum number of results to return in Search Points operation. Must be at least 1. |
Output
The node outputs an array of JSON objects corresponding to the result of the selected operation:
- For Get Collection Info, the output JSON contains detailed metadata about the specified collection.
- For Search Points, it returns search results including matched points and their scores.
- For List Collections, it outputs an array of all existing collections.
- For Count Points, it returns the count of points matching the filter.
- For Create/Update/Delete Collection and Upsert/Update/Delete Points, it returns success confirmation or the updated resource details.
- For Get Points by IDs, it returns the requested points.
No binary data output is produced by this node.
Dependencies
- Requires an active Qdrant API endpoint URL and an API key credential for authentication.
- Uses the official Qdrant JavaScript client library (
@qdrant/js-client-rest). - Requires proper n8n credentials setup with the Qdrant API URL and API key.
Troubleshooting
- Missing Collection Name: Most operations require a collection name; ensure this property is set and not empty.
- Invalid JSON Inputs: Properties like Collection Config, Point IDs, Points, Search Vector, and Filter expect valid JSON. Invalid JSON strings will cause errors.
- Required Parameters Missing: Operations like
createCollection,updateCollection,getPoints,deletePoints,upsertPoints,searchPointsrequire specific parameters; missing these will throw errors. - Limit Value: For
searchPoints, the limit must be at least 1; otherwise, an error is thrown. - API Errors: Network issues, invalid API keys, or incorrect endpoint URLs will cause API errors. Verify credentials and connectivity.
- Unsupported Operation: If an unsupported operation value is provided, the node throws an error indicating so.