mcpField
V
vector-index-tuning

vector-index-tuning

github.com/wshobsonData & databases41.9k10.8k installsUpdated 2w ago
Repository

About vector-index-tuning

A vector index tuning guide for retrieval engineers. It explains index choices, HNSW parameters, and quantization options to help balance recall, latency, and memory use.

Key features

  1. 01Index type selection guidance
  2. 02HNSW parameter explanations
  3. 03Quantization type comparisons
  4. 04Templates and detailed examples
  5. 05Production tuning practices

Use cases

  1. 01Reduce vector search latency
  2. 02Lower index memory use
  3. 03Balance retrieval recall and speed
  4. 04Plan large-scale vector indexes

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/wshobson/agents --skill vector-index-tuning

After installing, type/vector-index-tuningin Claude Code to confirm it loaded.

Bundled filesreferences/details.md

Details

From the project README

Vector Index Tuning

Guide to optimizing vector indexes for production performance.

When to Use This Skill

  • Tuning HNSW parameters
  • Implementing quantization
  • Optimizing memory usage
  • Reducing search latency
  • Balancing recall vs speed
  • Scaling to billions of vectors

Core Concepts

1. Index Type Selection
Data Size           Recommended Index
────────────────────────────────────────
< 10K vectors  →    Flat (exact search)
10K - 1M       →    HNSW
1M - 100M      →    HNSW + Quantization
> 100M         →    IVF + PQ or DiskANN
2. HNSW Parameters
ParameterDefaultEffect
M16Connections per node, ↑ = better recall, more memory
efConstruction100Build quality, ↑ = better index, slower build
efSearch50Search quality, ↑ = better recall, slower search
3. Quantization Types
Full Precision (FP32): 4 bytes × dimensions
Half Precision (FP16): 2 bytes × dimensions
INT8 Scalar:           1 byte × dimensions
Product Quantization:  ~32-64 bytes total
Binary:                dimensions/8 bytes

Templates and detailed worked examples

Full template library and detailed worked examples live in references/details.md. Read that file when you need the concrete templates.

Best Practices

Do's
  • Benchmark with real queries - Synthetic may not represent production
  • Monitor recall continuously - Can degrade with data drift
  • Start with defaults - Tune only when needed
  • Use quantization - Significant memory savings
  • Consider tiered storage - Hot/cold data separation
Don'ts
  • Don't over-optimize early - Profile first
  • Don't ignore build time - Index updates have cost
  • Don't forget reindexing - Plan for maintenance
  • Don't skip warming - Cold indexes are slow

Comments

Markdown supported