About vector stores

Vector stores are specially designed databases used to store a one-dimensional array of numerical data used to map bias and weight in machine learning models.

Vector stores require three connections:

• An embedding model node
• A vector retriever node
• A Large Language Model or Chat Model node

Before you begin

• Learn about vector stores in machine learning
• Learn about pathways
• Learn about pathway nodes

What vector stores are available for my pathway?

Node	Description	Additional information
FeatureBase Vector store	A FeatureBase table that uses the Vector data type	FeatureBase vector store
Vector store retriever	Retrieve data from a vector store to use in a pathway	Vector store retriever

What is a Vector store retriever?

The vector store retriever node obtains data from the vector store so it