VectorDB

A VectorDB is a specialized database storing high-dimensional vector embeddings to power rapid, semantic similarity search for AI applications.

Vector Databases (VectorDBs) are purpose-built systems designed to index and manage high-dimensional numerical arrays, or embeddings, which represent unstructured data like text, images, or audio. These vectors capture the semantic meaning of the data, allowing the database to perform efficient Approximate Nearest Neighbor (ANN) searches using algorithms like HNSW (Hierarchical Navigable Small World). This capability is critical for modern AI, directly enabling use cases like Retrieval-Augmented Generation (RAG) to ground Large Language Models (LLMs) in specific knowledge, and powering recommendation systems (e.g., Netflix) and semantic search. Key platforms in this space include Pinecone, Milvus, and Qdrant.

https://en.wikipedia.org/wiki/Vector_database