Glossary · Generative AI and LLMs
What is FAISS? Definition and meaning
Meta's library for nearest-neighbour search and clustering of dense vectors, with CPU and GPU indexes.
Definition
FAISS: Meta's library for nearest-neighbour search and clustering of dense vectors, with CPU and GPU indexes.
In short: Library for efficient similarity search.
When it is used
The index layer for custom retrieval and embedding research.
Where it fits
FAISS belongs to Vector databases in the Generative AI and LLMs topic. These are the other terms in the same group:
- Weaviate: A vector database that stores objects with embeddings and supports hybrid and filtered search.
- ChromaDB: A simple embedding store that is easy to embed in an application.
- Milvus: A distributed vector database built for billion-scale similarity search.
- Qdrant: A vector database with strong filtering, payloads and efficient indexing.
- Vespa: An engine that combines text, vector and structured search with on-node ranking.
- pgvector: A PostgreSQL extension that stores embeddings and supports exact and approximate nearest-neighbour search.
Learn more
The full guide to FAISS explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Experience replay · Next term: Feature engineering