Qdrant icon

Qdrant

Official n8n node to interface with Qdrant - https://qdrant.tech

Actions27

Search → Query Points

AI-generated

Overview

This node operation allows querying points from a specified collection in Qdrant, a vector search engine. It supports various query types including vector queries, ID lookups, and complex query objects like recommend, fusion, sample, and formula. Users can filter results, set score thresholds, limit the number of results, and control the inclusion of vectors and payloads in the response. This operation is useful for applications requiring advanced vector search capabilities, such as recommendation systems, similarity searches, and data retrieval based on vector embeddings.

Use Case Examples

  1. Querying a collection to find the top 50 points most similar to a given vector embedding.
  2. Filtering points in a collection based on specific payload conditions and retrieving their associated data.
  3. Performing a batch query with prefetch to optimize search results within a subset of points.

Properties

Name Meaning
Collection Name Specifies the name of the collection to query points from. It can be selected from a list or entered by name.
Query Defines the query to perform, which can be a vector, an ID, or a complex query object such as recommend, fusion, sample, or formula.
Score Threshold Filters points to return only those with scores better than this threshold.
Limit Sets the maximum number of results to return from the query.
Offset Specifies the number of points to skip in the result set, useful for pagination.
With Vector Determines whether to include the vector data in the response.
With Payload Determines whether to include the payload data in the response.
Using Specifies which vector name to use for querying if multiple vectors exist.
Filter Applies conditions to select only points that satisfy these criteria.
Prefetch Defines sub-requests to perform first; the main query is then performed on the results of these prefetches.
Params Search parameters used when there is no prefetch defined.
Lookup From Specifies the location to use for IDs lookup.
Consistency Defines the read consistency guarantees for the operation, such as majority, quorum, or all.
Timeout Overrides the global timeout for this request in seconds.
Shard Key Specifies which shards to look for points in; if not specified, all shards are searched.
Request Options Additional request settings such as batching, SSL certificate validation, proxy configuration, and request timeout.

Output

JSON

  • result
    • points - Array of points returned by the query, each containing the point data matching the query criteria.
    • count - The total count of points matching the query criteria.
    • status - Status of the query operation indicating success or failure.

Dependencies

  • Qdrant REST API

Troubleshooting

  • Ensure the collection name specified exists; otherwise, the query will fail.
  • Verify the query format is correct (vector, ID, or valid complex query object) to avoid errors.
  • Check network connectivity and API credentials if requests time out or fail.
  • If using filters or prefetch, ensure the JSON structure is valid and matches expected schema.
  • Adjust the score threshold and limit parameters to get meaningful results without overwhelming the response.

Links

Discussion