Vector Store Documentation

The Vector Store APIs allow you to create, manage, and search vector stores in Rememberizer. Vector stores enable you to store and retrieve documents using semantic similarity search.

Available Vector Store Endpoints

Management Endpoints

• Get vector store's information

• Get a list of documents in a Vector Store

• Get the information of a document

Document Operations

• Add new text document to a Vector Store

• Upload files to a Vector Store

• Update file's content in a Vector Store

• Remove a document in Vector Store

Search Operations

• Search for Vector Store documents by semantic similarity

Creating a Vector Store

To create a new Vector Store, use the following endpoint:

POST /api/v1/vector-stores/

Request Body

{
  "name": "Store name",
  "description": "Store description",
  "embedding_model": "sentence-transformers/all-mpnet-base-v2",
  "indexing_algorithm": "ivfflat",
  "vector_dimension": 128,
  "search_metric": "cosine_distance"
}

Response

{
  "id": "store_id",
  "name": "Vector Store Name",
  "description": "Store description",
  "created": "2023-05-01T00:00:00Z",
  "modified": "2023-05-01T00:00:00Z"
}

Vector Store Configurations

To retrieve available configurations for vector stores, use:

GET /api/v1/vector-stores/configs

This will return available embedding models, indexing algorithms, and search metrics that can be used when creating or configuring vector stores.

Authentication

All Vector Store endpoints require authentication using either:

• JWT token for management operations

• API key for document and search operations