Concepts

HIDDB is a vector database which is optimized towards efficient similarity search. The database is fully-managable, easy to integrate in existing projects and blazingly fast.

Similarity search and vector embeddings

Vector embeddings, are sets of floating-point numbers that represent objects, like images or audios. They are generated by embedding models trained to capture the semantic similarity of objects in a given set.

Unlike traditional search methods that revolve around keywords, HIDDB creates and indices to search through your vector embeddings to find similar items.

In order to use HIDDB you need to have vector embeddings of some kind.

HIDDB indexes and stores vector embeddings for efficient management and fast similarity search.

HIDDB datastructures

HIDDB stores data in documents which can be represented as JSON objects. Documents are stored in collections (similar to SQL tables but with no schema) within a database to store and organize documents. A database stores one or more collections of documents.

HIDDB is a database which is optimized towards efficient similarity search. Similarity search is realized via indices which are specific to a collection.

Working with HIDDB

A RESTful API and documentated here. Consider using the Python SDK or Javascript SDK.