pgvector
Vector similarity search as a PostgreSQL extension, keeping embeddings beside your data.
Visão geral
pgvector adds vector columns and approximate nearest-neighbour indexes to PostgreSQL. It solves the operational problem of a second datastore: when embeddings live in their own database, every query has to join across two systems and every transaction has to consider both. It exists because for most applications the vector count is small enough that one database is simpler, and filtering by ordinary columns alongside similarity is exactly where a dedicated vector database is weakest.
Grafo de tecnologias
Como pgvector se liga ao resto do stack.
Usa e depende de
- Embedding concept Techniques pgvector uses Embedding High ✓ source
Classification
- PostgreSQL database Databases pgvector part of PostgreSQL High ✓ source
Ferramentas
Software que implementa ou se integra com pgvector.