Verified:GitHub

embedding-strategies

Select and optimize embedding models and chunking strategies for RAG pipelines.

1|Updated Feb 19, 2026

Search Tags:#semantic search #embeddings #llm #rag #vector database #chunking #model optimization

One-click install

npx skills add https://github.com/yusufcmg/Antigravity-Agents-Workflows --skill embedding-strategies-yusufcmg

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: embedding-strategies
Source: https://github.com/yusufcmg/Antigravity-Agents-Workflows/tree/main/.agent/skills/ai/embedding-strategies
Command: npx skills add https://github.com/yusufcmg/Antigravity-Agents-Workflows --skill embedding-strategies-yusufcmg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires langchain-voyageai, openai, sentence-transformers, tiktoken, nltk, tree-sitter-languages, numpy, scipy, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you select, implement, and optimize embedding models and chunking strategies for efficient semantic search and Retrieval Augmented Generation (RAG) applications.

Core Features & Use Cases

Model Selection: Compare various embedding models (Voyage AI, OpenAI, Sentence Transformers) based on dimensions, cost, and best use cases (code, finance, general).
Chunking Strategies: Implement different methods for splitting text into manageable chunks (token-based, sentence-based, semantic sections, recursive).
Pipeline Implementation: Provides Python templates for integrating embedding models and chunking into your RAG pipeline.
Quality Evaluation: Includes functions to evaluate retrieval quality using metrics like Precision@K, Recall@K, MRR, and NDCG.

Quick Start

Use the embedding-strategies skill to get embeddings for the provided list of texts using the 'text-embedding-3-small' model.

Frequently Asked Questions about embedding-strategies

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

FAQPage Schema

How do I choose the best embedding model for a RAG pipeline?▼

Chunking strategies for semantic search split text into manageable pieces using token-based, sentence-based, semantic sections, or recursive methods, directly impacting retrieval quality and overall RAG pipeline performance.

Can I evaluate embedding retrieval quality using Python?▼

You can implement a semantic search pipeline using provided Python templates that integrate Voyage AI, OpenAI, or local Sentence Transformers models with your vector database for efficient text retrieval.

Does this approach work for generating embeddings from source code?▼

When optimizing embeddings for RAG, consider token limits using tiktoken and evaluate domain-specific chunking strategies, as recursive or semantic splitting may outperform standard token-based methods depending on your text structure.

What is the best way to chunk text for semantic search?▼

The best way to chunk text for semantic search depends on your content, utilizing token-based, sentence-based, recursive, or semantic section splitting to maximize retrieval quality within your vector database.