skills.rest
afel6
afel6COMMUNITY
Explore all skills by this author →
GitHub

faiss

Build and query FAISS indexes for vector similarity search over dense embeddings.

Updated Mar 25, 2026
Search Tags:#indexing#gpu#similarity#knn#embedding#vector-search#faiss
One-click install
npx skills add https://github.com/afel6/scal-ai-pipeline --skill faiss-afel6
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: faiss
Source: https://github.com/afel6/scal-ai-pipeline/tree/main/hermes_skills_library/optional-skills/mlops/faiss
Command: npx skills add https://github.com/afel6/scal-ai-pipeline --skill faiss-afel6

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires faiss-cpu, faiss-gpu, numpy, and includes references (resource) components.

What problem does it solve?

FAISS provides fast, scalable vector similarity search across massive embedding collections, enabling instant nearest neighbor queries for large-scale retrieval tasks.

Core Features & Use Cases

  • Large-scale vector search with CPU and GPU acceleration
  • Multiple index types: Flat, IVF, HNSW, PQ for trade-offs between speed and memory
  • Use cases: retrieval-augmented generation, large-scale recommendation, and similarity-based filtering

Quick Start

Install FAISS, create an index, add vectors, and run a search on your dataset.

Frequently Asked Questions about faiss

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform fast similarity search over billions of dense embeddings?▼

Fast similarity search over billions of dense embeddings is achieved by building and querying FAISS indexes, which provide low latency k-NN results for large-scale retrieval tasks. It supports CPU and optional GPU acceleration.

Can I use GPU acceleration for k-NN vector search?▼

Yes, GPU acceleration for k-NN vector search is supported through the faiss-gpu dependency, enabling faster similarity queries across massive embedding collections alongside standard CPU processing.

What's the best way to balance speed and memory for large-scale vector search?▼

Balancing speed and memory for large-scale vector search is done by selecting among multiple FAISS index types like Flat, IVF, HNSW, and PQ, each offering distinct trade-offs for your specific retrieval scale.

Do I need numpy to build a vector similarity search index?▼

Yes, numpy is required alongside faiss-cpu or faiss-gpu to build and query vector similarity indexes, as it handles the dense embedding array inputs necessary for the indexing process in Python environments.

When should I use FAISS for retrieval augmented generation?▼

You should use FAISS for retrieval augmented generation when you need to quickly find relevant context from massive embedding collections, providing instant nearest neighbor queries to feed downstream generation models.

Does FAISS support offline analytics for similarity-based filtering?▼

Yes, FAISS supports offline analytics and similarity-based filtering by running batch k-NN queries over large datasets, enabling scalable recommendation and retrieval tasks without real-time latency constraints.

Related Skills

Explore More Data & Analytics Skills→
GregsGreyCodeGregsGreyCode
6

faiss

Perform scalable vector similarity search over embeddings with FAISS indices.

Community
Intermediate
samuelmukotisamuelmukoti

faiss

Perform vector similarity search with FAISS index types and GPU acceleration.

Community
Advanced
AnandaAnugrahHandyantoAnandaAnugrahHandyanto

faiss

Build and search FAISS indexes for large-scale vector similarity.

Community
Advanced
ever-oliever-oli
3

faiss

Create FAISS indexes and run k-nearest-neighbor queries on large embeddings.

Community
Advanced