intermediateAnalysis & Research

Vector Index Tuning

Optimize vector index performance for latency, recall, and memory. Use when tuning HNSW parameters, selecting quantization strategies, or scaling vector search infrastructure.

View source
84OpxScoreProvisional
Community resultNot enough feedback0 votes
Model evidenceNo verified tests
ClaudeChatGPTGemini+5

Score breakdown

Estimated from the available content and source signals.

Provisional
Documentation92
Practical value92
Evidence63
Source trust80

Model compatibility

Inferred fit is not the same as a recorded hands-on test.

ClaudeinferredThe skill uses model-agnostic prompt or LLM terminology.
ChatGPTinferredThe skill uses model-agnostic prompt or LLM terminology.
GeminiinferredThe skill uses model-agnostic prompt or LLM terminology.
CopilotinferredThe skill uses model-agnostic prompt or LLM terminology.
LlamainferredThe skill uses model-agnostic prompt or LLM terminology.
PerplexityinferredThe skill uses model-agnostic prompt or LLM terminology.
MistralinferredThe skill uses model-agnostic prompt or LLM terminology.
GrokinferredThe skill uses model-agnostic prompt or LLM terminology.

Overview

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

Best for

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

Tips and best practices

  • Review the source instructions and adapt inputs before running the workflow.

What This Skill Can Do

AI-generated examples showing real capabilities

Was this skill useful?

Be the first to share a result.

Related skills