Actions26
- Collection Actions
- Payload Actions
- Point Actions
- Search Actions
- Vector Actions
Overview
The node provides an interface to query points from a vector search engine collection. It allows users to perform complex vector similarity searches or ID-based lookups within a specified collection. This is useful in scenarios such as semantic search, recommendation systems, or any application requiring fast retrieval of items similar to a given vector or matching certain criteria.
For example, you can use this node to:
- Find the top 50 most similar documents to a given embedding vector.
- Retrieve points with a score above a threshold to filter out less relevant results.
- Include additional metadata (payload) or vectors in the response for richer context.
- Apply filters to narrow down search results based on custom conditions.
Properties
| Name | Meaning |
|---|---|
| Collection Name | The target collection to query points from. Can be selected from a list of existing collections or specified by name. |
| Query | The query to perform. Can be a vector (array of numbers), a point ID, or a complex query object supporting recommend, fusion, sample, or formula queries. |
| Score Threshold | Minimum score threshold; only points with scores better than this value will be returned. |
| Limit | Maximum number of results to return. Must be at least 1. |
| Offset | Number of points to skip before starting to collect results (useful for 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 | Specifies which named vector field to use for querying if multiple vectors exist per point. |
| Filter | JSON object defining conditions to filter points before querying. |
| Prefetch | JSON object specifying sub-requests to perform first; the main query will run on the prefetch results. |
| Params | Additional search parameters used when no prefetch is specified. |
| Lookup From | JSON object defining where to lookup IDs from 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 object specifying which shards to search in; if omitted, all shards are searched. |
Output
The output contains a JSON array of points matching the query criteria. Each point typically includes:
- An identifier.
- A similarity score relative to the query.
- Optionally, the vector data if
With Vectoris true. - Optionally, the payload metadata if
With Payloadis true.
If binary data were involved (not indicated here), it would represent raw vector or payload content, but this node focuses on JSON structured data.
Dependencies
- Requires an API key credential for authenticating requests to the Qdrant REST API.
- Needs the base URL of the Qdrant service configured in credentials.
- No other external dependencies are indicated.
Troubleshooting
Common issues:
- Invalid or missing collection name: Ensure the collection exists and is correctly referenced.
- Malformed query JSON: Validate the query structure matches expected formats (vector array, ID, or complex query object).
- Timeout errors: Increase the timeout property if large queries take longer.
- Insufficient permissions or invalid API key: Verify credentials are correct and have access rights.
Error messages:
- "Collection not found": Check the collection name spelling and existence.
- "Invalid query format": Review the query JSON syntax and semantics.
- "Timeout exceeded": Adjust the timeout setting or optimize the query.