Faiss

FAISS is Meta's battle-tested vector similarity search library, and it's what you reach for when you need to handle millions or billions of vectors without everything grinding to a halt. It's pure similarity search, no metadata filtering or fancy database features, just blazing fast k-NN powered by C++ with GPU support that can give you 10-100× speedups. The index variety is solid: Flat for exact search, IVF for speed, HNSW for best quality-to-speed ratio, and PQ when memory gets tight. It integrates cleanly with LangChain and LlamaIndex, and with 31,700+ GitHub stars, you're not exactly beta testing here. If you need metadata filtering alongside search, grab Chroma or Pinecone instead. For pure vector retrieval at scale, this is the standard.

skills.sh installs: 1.1K
GitHub stars: 14.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add orchestra-research/ai-research-skills --skill faiss --agent claude-code

Installs into .claude/skills of the current project.

Files

FAISS - Efficient Similarity Search

Facebook AI's library for billion-scale vector similarity search.

When to use FAISS

Use FAISS when:

Need fast similarity search on large vector datasets (millions/billions)
GPU acceleration required
Pure vector similarity (no metadata filtering needed)
High throughput, low latency critical
Offline/batch processing of embeddings

Metrics:

31,700+ GitHub stars
Meta/Facebook AI Research
Handles billions of vectors
C++ with Python bindings

Use alternatives instead:

Chroma/Pinecone: Need metadata filtering
Weaviate: Need full database features
Annoy: Simpler, fewer features

Quick start

Installation

# CPU only
pip install faiss-cpu

# GPU support
pip install faiss-gpu

Basic usage

import faiss
import numpy as np

# Create sample data (1000 vectors, 128 dimensions)
d = 128
nb = 1000
vectors = np.random.random((nb, d)).astype('float32')

# Create index
index = faiss.IndexFlatL2(d)  # L2 distance
index.add(vectors)             # Add vectors

SKILL.md on GitHub