Back to the AI glossary Infrastructure
Vector database
Database optimized for quick searches for similar content.
Explanation
A vector database stores embeddings and enables extremely fast similarity searches. It is the central storage system for RAG applications and semantic search.
How it works
Texts are saved as embedding vectors. When a search query is made, the database calculates which stored vectors are most similar to the query - in milliseconds.
Example
Pinecone, Weaviate or Qdrant as a vector database for a company chatbot that can search thousands of internal documents.
Why it matters
RAG would not be possible without a vector database. They are the infrastructure behind every AI that works with its own data.