FAISS — Fast Vector Similarity Search

FAISS (Facebook AI Similarity Search) is a library for efficient similarity search over large collections of vectors. No server required — runs entirely in-process.

When to Use FAISS

• Millions of vectors, need maximum speed
• Can't run a server (embedded use case)
• Research / prototyping at scale
• Custom index types (HNSW, IVF, PQ)

Setup

pip install faiss-cpu        # CPU only
pip install faiss-gpu        # GPU (requires CUDA)
pip install sentence-transformers numpy