skills.rest
ACGSpgp
ACGSpgpOFFICIAL
Explore all skills by this author →
WebsiteGitHub

vector-index-tuning

Tune HNSW parameters and quantization strategies for vector database performance.

Updated Aug 23, 2026
Search Tags:#vector database#performance tuning#similarity search#quantization#hnsw#vector index
One-click install
npx skills add https://github.com/ACGSpgp/ACGS --skill vector-index-tuning-acgspgp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vector-index-tuning
Source: https://github.com/ACGSpgp/ACGS/tree/main/.agent/workflows/llm-application-dev/1.2.1/skills/vector-index-tuning
Command: npx skills add https://github.com/ACGSpgp/ACGS --skill vector-index-tuning-acgspgp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires hnswlib, scikit-learn, qdrant-client, numpy, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the critical challenge of optimizing vector index performance, ensuring efficient and accurate similarity searches in large-scale vector databases.

Core Features & Use Cases

  • HNSW Parameter Tuning: Adjust M, efConstruction, and efSearch for optimal recall and latency.
  • Quantization Strategies: Implement INT8, Product Quantization, or Binary quantization to reduce memory footprint.
  • Use Case: When deploying a new recommendation system with millions of user embeddings, use this Skill to find the best HNSW parameters and quantization method to achieve sub-100ms search latency with over 95% recall.

Quick Start

Use the vector-index-tuning skill to recommend HNSW parameters for 1 million vectors targeting 95% recall.

Frequently Asked Questions about vector-index-tuning

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

FAQPage Schema
How do I tune HNSW parameters for vector search to improve recall and latency?▼

Tuning HNSW parameters involves adjusting M, efConstruction, and efSearch to balance graph connectivity and search depth, directly optimizing vector search recall and latency for large-scale similarity search.

What's the best way to reduce memory footprint in a vector database without losing search accuracy?▼

Reducing vector database memory footprint requires applying quantization strategies like INT8, Product Quantization, or Binary quantization, which compress vector embeddings while maintaining high similarity search recall.

Can I achieve sub-100ms similarity search latency with 95% recall on 1 million vectors?▼

Achieving sub-100ms similarity search latency with 95% recall on 1 million vectors is possible by optimizing HNSW configurations and selecting appropriate quantization methods for the vector index.

Does this vector index optimization approach work with hnswlib and qdrant-client?▼

This vector index optimization approach supports hnswlib and qdrant-client environments, utilizing numpy and scikit-learn to evaluate and tune vector database performance for large-scale similarity search.

Why does my vector search latency spike when scaling to large datasets?▼

Vector search latency spikes during large dataset scaling often occur due to unoptimized HNSW parameters or high memory consumption, requiring index tuning and quantization strategies to restore efficient performance.

Related Skills

Explore More Software Engineering Skills→
andikarachmanandikarachman
14

data-preprocessing

Preprocess raw data into clean, analysis-ready datasets with pandas and numpy.

Community
Advanced
EngineerWithAIEngineerWithAI
4

vector-index-tuning

Tune HNSW parameters and quantization strategies for Qdrant vector indexes.

Official
Advanced
JuanMS20JuanMS20

qdrant-vector-search

Enable vector search and semantic retrieval with Qdrant.

Community
Advanced
DrLuggelsDrLuggels
3

vector-index-tuning

Optimize vector index performance for latency, recall, and memory usage.

Community
Advanced