Vector database

A vector database is a type of database ✕ A database stores a site’s content —posts, pages, comments— in an organized way so it can be retrieved instantly. WordPress uses MySQL or MariaDB. More in the glossary → designed to store embeddings ✕ An embedding turns a piece of data — a word, a sentence, an image — into a list of numbers that captures its meaning, so that similar things end up close together in that numeric space. More in the glossary → — those lists of numbers that represent the meaning of texts, images or other data — and search them by similarity. Instead of looking for exact matches like a traditional database, it finds, given a query vector, the vectors "nearest" to it, that is, the items closest in meaning. This makes semantic search possible at scale and very fast, even with millions of items. It is a key piece of many modern AI ✕ Artificial intelligence is a computer system’s ability to perform tasks we associate with human intelligence, such as understanding language, recognizing images or making decisions. More in the glossary → applications: it powers recommendation systems and, above all, retrieval-augmented generation ✕ RAG connects a language model to an external data source, so it answers based on real, up-to-date information instead of only its memory. More in the glossary → (RAG), in which a language model retrieves relevant information before answering. Well-known examples are Pinecone, Weaviate or PostgreSQL’s pgvector extension. Its rise goes hand in hand with that of generative AI ✕ Generative AI creates new content —text, images, audio or code— from a prompt, instead of only analyzing existing data. More in the glossary → .