Qdrant (Beta)
The Qdrant Micro-Integration publishes events from a PubSub+ event broker to a Qdrant vector database. It converts event data into vectors using an embedding model. A vector is a list of numbers representing a concept or semantic meaning of a chunk of text. An embedding model is a machine learning model that performs the data conversion.
You can configure Qdrant only as a target.
Qdrant Target Parameters
The following table describes the parameters for configuring Qdrant as a target.
| Field | Description |
|---|---|
| Qdrant Target Connection Details | |
|
Select Content Splitter |
Select the method for splitting incoming content for vectorization. Valid values are:
|
| OpenAI-Compatible Embedding Model Provider | |
|
Endpoint URL |
The API endpoint URL for the embedding model provider. For example: |
|
Model ID |
The specific version or name of the embedding model to use, following the The access protocol provider specifies which service to use to access the model. The model ID specifies the embedding model within that provider's catalog. The value of |
|
Secret Access Key |
The API Secret Access Key for the embedding model provider. |
|
Organization ID |
The OpenAI Organization ID. |
|
Chunk Size |
The size of text segments into which a document is divided by an embedding model for vector database storage. The default is 1000. |
|
Embedding Model Context Length |
The maximum number of tokens to embed at once by the embedding model. The default is 8191. |
|
Number of Embedding Dimensions |
The number of dimensions in the resulting output embeddings. |
| Qdrant Vector Database | |
|
Endpoint URL |
The API endpoint URL for the Qdrant Vector Database. For example: |
|
Secret Access Key |
The API Secret Access Key for the Qdrant provider. |
|
Collection Name |
The name of the Qdrant collection. A collection in Qdrant is a container that stores vectors where all vectors must have the same dimensions and have used the same model for their creation. Collections allow you to efficiently search for and retrieve vectors that are similar to a vector you specify in a query. For more information, see Collections in the Qdrant documentation. If the specified collection does not exist, the Micro-Integration creates a new empty basic collection with the specified name. |
JSON Document Splitter Parameters
| Field | Description |
|---|---|
|
Max Chunk Size |
The maximum number of tokens per text chunk during document splitting. Controls the size of segments created during text ingestion for vector storage. The default is 2000. Larger chunks preserve more context but may reduce retrieval precision. Smaller chunks improve granular search but may lose contextual relationships. |
Character Document Splitter Parameters
| Field | Description |
|---|---|
|
Chunk Size |
The maximum number of characters allowed in each text segment when splitting a document. The default is 4000. |
|
Chunk Overlap |
The number of characters that overlap between consecutive chunks. Overlap helps to maintain context between split segments. The default is 200. |
|
Separator |
The character sequence used to split text into chunks, allowing control over split points at natural boundaries like paragraph breaks. The default is |
Structured Text Splitter Parameters
| Field | Description |
|---|---|
|
Chunk Size |
The maximum number of characters allowed in each text segment when splitting a document. The default is 4000. |
|
Chunk Overlap |
The number of characters that overlap between consecutive chunks. Overlap helps to maintain context between split segments. The default is 200. |
|
Separators |
The character sequence used to split text into chunks based on text structure allowing control over split points at natural boundaries like paragraph breaks. This can be a comma-separated list of separators like: |