skills.rest
Rigohl
RigohlCOMMUNITY
Explore all skills by this author →
GitHub

faiss-gpu-optimization

Configure FAISS-GPU with IVF-PQ indexing for billion-scale vector search.

Updated Dec 30, 2025
Search Tags:#indexing#embeddings#gpu#cuda#vector-search#faiss
One-click install
npx skills add https://github.com/Rigohl/MEMORY_P --skill faiss-gpu-optimization
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: faiss-gpu-optimization
Source: https://github.com/Rigohl/MEMORY_P/tree/main/.github/skills/faiss-gpu-optimization
Command: npx skills add https://github.com/Rigohl/MEMORY_P --skill faiss-gpu-optimization

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

FAISS-GPU acceleration enables ultra-fast vector search on NVIDIA GPUs for billion-scale datasets, dramatically reducing latency for large-scale similarity search workloads.

Core Features & Use Cases

  • GPU-accelerated FAISS vector search with IVF-PQ indexing for massive datasets
  • Support for multi-GPU deployment and production search pipelines
  • Memory-aware indexing and batch processing to maximize throughput
  • Production-ready search engine integration with batch and streaming queries

Quick Start

Query a trained FAISS-GPU index with a batch of embeddings to retrieve the top-k most similar items.

Frequently Asked Questions about faiss-gpu-optimization

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

FAQPage Schema
How do I run billion-scale vector search on GPU?▼

Billion-scale vector search on GPU uses FAISS-GPU with IVF-PQ indexing and memory-optimized pipelines to accelerate similarity search. It trains, indexes, and queries large vector collections on CUDA-enabled NVIDIA GPUs to dramatically reduce latency for high-throughput workloads.

Can I use FAISS-GPU for multi-GPU deployment?▼

FAISS-GPU supports multi-GPU deployment for production search pipelines. It applies memory-aware indexing and batch processing across multiple CUDA-enabled NVIDIA GPUs to maximize throughput for large-scale similarity search workloads.

Do I need CUDA to run GPU-accelerated FAISS vector search?▼

GPU-accelerated FAISS vector search requires CUDA-enabled NVIDIA GPUs and the CUDA toolkit. A Python environment is also necessary to train, index, and query large vector collections using IVF-PQ indexing.

What's the best way to handle memory limits in billion-scale vector search?▼

Memory-aware indexing and batch processing handle memory limits in billion-scale vector search. By applying IVF-PQ indexing with FAISS-GPU, you can compress vector collections and maximize throughput on CUDA-enabled NVIDIA GPUs.

How does IVF-PQ indexing work for high-throughput similarity search?▼

IVF-PQ indexing partitions vector collections into clusters and compresses embeddings into shorter codes for high-throughput similarity search. FAISS-GPU accelerates this process on CUDA-enabled NVIDIA GPUs, enabling ultra-fast querying of billion-scale datasets.

How do I query a trained FAISS-GPU index with batch embeddings?▼

You query a trained FAISS-GPU index by passing a batch of embeddings through a Python environment to retrieve the top-k most similar items. This uses memory-optimized pipelines on CUDA-enabled NVIDIA GPUs for fast similarity search results.

Related Skills

Explore More Software Engineering Skills→
ever-oliever-oli
3

tensorrt-llm

Optimize LLM inference on NVIDIA GPUs with TensorRT and CUDA graphs.

Community
Advanced
KarlinskySKarlinskyS

faiss

Perform fast nearest-neighbor search on dense vectors using FAISS indexes.

Community
Intermediate
zechenzhangAGIzechenzhangAGI
11.5k

faiss

Perform billion-scale k-NN similarity search with FAISS index types and GPU acceleration.

Community
Intermediate
blueskies1818blueskies1818
1

faiss

Perform k-NN vector similarity search and clustering with FAISS.

Community
Advanced