---
id: llama-index-vector-stores-faiss
version: "0.6.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# llama-index-vector-stores-faiss — llama-index vector_stores faiss integration

License: permissive · Maintenance: active · Downloads: 150.2K/mo

## What it is and what it does

This package provides a LlamaIndex integration for FAISS, a library for efficient similarity search over high-dimensional vectors. It acts as a bridge between LlamaIndex's document and embedding management layer and FAISS's vector indexing capabilities, allowing you to store embeddings and perform fast nearest-neighbor queries within a LlamaIndex application.

The package is designed for developers building retrieval-augmented generation (RAG) systems, semantic search applications, or other AI workflows that need to index and query embeddings. It depends only on llama-index-core and installs with low friction. The MIT license and active maintenance status make it a straightforward choice for integrating FAISS into LlamaIndex-based projects.

Use it for:

- Store and retrieve document embeddings in retrieval-augmented generation (RAG) pipelines using LlamaIndex.
- Build semantic search applications that find similar documents or passages based on embedding similarity.
- Integrate efficient in-memory or disk-backed vector indexing into LlamaIndex workflows without external database infrastructure.
- Prototype and develop AI applications that need fast approximate nearest-neighbor search over embeddings.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Integrates FAISS vector storage with LlamaIndex for efficient similarity search and retrieval in RAG applications.

Yes. This is a lightweight, actively maintained integration package with no known vulnerabilities, permissive licensing, and low installation friction. Install it if you are building a LlamaIndex application and want to use FAISS for vector storage and retrieval.

## Install

pip install llama-index-vector-stores-faiss
uv add llama-index-vector-stores-faiss
poetry add llama-index-vector-stores-faiss

## Installing llama-index-vector-stores-faiss

Before you install: Low friction installation with a single runtime dependency on llama-index-core. Active maintenance status with recent release activity.

License in practice: MIT license permits use, modification, and distribution with minimal restrictions, making it suitable for both open-source and commercial projects.

Quickstart:

pip install llama-index-vector-stores-faiss

from llama_index.vector_stores.faiss import FaissVectorStore

vector_store = FaissVectorStore()

Requires Python 3.10 or later; FAISS may have platform-specific build requirements.

Verify before relying:

- Whether FAISS itself requires system-level dependencies or compilation on target platforms
- Performance characteristics and scalability limits for different index sizes
- Specific LlamaIndex version compatibility beyond the Python version requirement

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 150.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

faiss vector store integration, llama index vector storage, semantic search with faiss, rag vector database, similarity search integration, embedding storage and retrieval, llama index faiss connector, vector-database, rag, semantic-search

[View on SkillFed](https://skillfed.io/packages/llama-index-vector-stores-faiss) · [View on PyPI](https://pypi.org/project/llama-index-vector-stores-faiss/)