Redis Vector Store

​

Node Details

• Name: Redis_VectorStores

• Type: Redis

• Version: 1.0

• Category: Vector Stores

​

Base Classes

• Redis

• VectorStoreRetriever

• BaseRetriever

​

Parameters

​

Credential (Required)

• Label: Connect Credential

• Name: credential

• Type: credential

• Credential Names: redisCacheUrlApi, redisCacheApi

​

Inputs

1. Document (Optional)

   • Type: Document

   • List: true

   • Description: The documents to be inserted into the vector store.

2. Embeddings (Required)

   • Type: Embeddings

   • Description: The embedding model to use for converting documents into vectors.

3. Index Name (Required)

   • Type: string

   • Placeholder: VECTOR_INDEX_NAME

   • Description: The name of the Redis index to use.

4. Replace Index on Upsert (Optional)

   • Type: boolean

   • Default: false

   • Description: If true, deletes the existing index and recreates a new one when upserting.

5. Content Field (Optional)

   • Type: string

   • Default: “content”

   • Description: Name of the field that contains the actual content.

6. Metadata Field (Optional)

   • Type: string

   • Default: “metadata”

   • Description: Name of the field that contains the metadata of the document.

7. Vector Field (Optional)

   • Type: string

   • Default: “content_vector”

   • Description: Name of the field that contains the vector.

8. Top K (Optional)

   • Type: number

   • Default: 4

   • Description: Number of top results to fetch.

​

Outputs

1. Redis Retriever

   • Name: retriever

   • Base Classes: [Redis, VectorStoreRetriever, BaseRetriever]

2. Redis Vector Store

   • Name: vectorStore

   • Base Classes: [Redis, RedisVectorStore base classes]

​

Functionality

The node provides two main functionalities:

1. Upsert: Allows inserting or updating documents in the Redis vector store.

2. Similarity Search: Enables performing similarity searches on the stored vectors.

​

Usage

This node is typically used in workflows where fast similarity search is required on a large number of embedded documents. It’s particularly useful in applications such as:

• Semantic search engines

• Recommendation systems

• Document retrieval systems

• Any application requiring quick similarity comparisons between vectors

​

Implementation Details

• The node uses the redis npm package for Redis client operations.

• It implements a singleton pattern for the Redis client to optimize connections.

• The node handles both direct Redis URL connections and credential-based connections.

• It includes methods for upserting documents and performing similarity searches.

• The implementation includes error handling and index existence checks.

​

Note

Ensure that you are running a RediSearch-capable Redis instance for full functionality of this node.