Actions26
- Delete Collection
- Get Collection
- Check Collection Exists
- Create Collection
- Update Collection
- Retrieve Point
- Retrieve Points
- Upsert Points
- Delete Points
- Update Vectors
- Delete Vectors
- Set Payload
- Overwrite Payload
- Delete Payload
- Clear Payload
- Batch Update Points
- Scroll Points
- Count Points
- Query Points
- Query Batch Points
- Query Points Groups
- Matrix Pairs
- Matrix Offsets
- Payload Facets
- Create Payload Index
- Delete Payload Index
Overview
The node provides an interface to query points from a Qdrant vector search engine collection. It allows users to perform complex queries on stored vectors and their associated payloads, enabling retrieval of relevant data points based on vector similarity or other filtering criteria.
This node is beneficial in scenarios where you want to search for similar items in a vector database, such as:
- Finding nearest neighbors for recommendation systems.
- Retrieving documents or images similar to a query vector.
- Filtering points by metadata (payload) conditions combined with vector similarity.
- Implementing semantic search over large datasets.
For example, you could use this node to query a collection of product embeddings to find the top 50 products most similar to a given vector, optionally including their metadata and vectors in the response.
Properties
| Name | Meaning |
|---|---|
| Collection Name | The name of the Qdrant collection to query points from. |
| Query | The query object specifying what to search for. Can be a vector, point ID, or a complex query object supporting recommend, fusion, sample, formula types. |
| Score Threshold | Return only points with scores better than this threshold. |
| Limit | Maximum number of points to return. |
| Offset | Number of points to skip before starting to return results (pagination). |
| With Vector | Whether to include the vector data of each point in the response. |
| With Payload | Whether to include the payload (metadata) of each point in the response. |
| Using | Specify which named vector field to use for querying if multiple vectors exist per point. |
| Filter | JSON object defining conditions to filter points by their payload or other attributes. |
| Prefetch | JSON object defining sub-queries to perform first; the main query will run on the results of these prefetches. |
| Params | Additional search parameters to apply when there is no prefetch. |
| Lookup From | JSON object specifying the location to use for looking up IDs during the query. |
| Consistency | Read consistency level for the operation. Options: Majority, Quorum, All. |
| Timeout | Request timeout in seconds, overriding the global default. |
| Shard Key | JSON specifying which shards to look into; if not set, all shards are searched. |
Output
The output contains a JSON array of points matching the query criteria. Each point includes:
- Its unique identifier.
- Optionally, its vector data if "With Vector" is enabled.
- Optionally, its payload (metadata) if "With Payload" is enabled.
- A score indicating similarity or relevance according to the query.
If binary data were involved, it would represent raw vector bytes or serialized payloads, but this node primarily outputs structured JSON data representing points.
Dependencies
- Requires connection to a Qdrant vector search engine instance.
- Needs an API key credential configured in n8n for authentication.
- The base URL of the Qdrant service must be provided in credentials.
- The node uses HTTP requests with JSON payloads to communicate with Qdrant.
Troubleshooting
Common issues:
- Incorrect collection name causing "collection not found" errors.
- Malformed query JSON leading to request failures.
- Timeout errors if the query takes too long; increase the timeout property.
- Insufficient permissions or invalid API key causing authentication errors.
Error messages:
- "Collection does not exist": Verify the collection name is correct and exists in Qdrant.
- "Invalid query format": Check the syntax and structure of the query JSON.
- "Request timed out": Increase the timeout value or optimize the query.
- "Unauthorized" or "Forbidden": Confirm API key validity and permissions.