selecting-embeddings

What problem does it solve?

This Skill helps you select and optimize the most suitable embedding models for your specific semantic search and Retrieval Augmented Generation (RAG) applications, ensuring better accuracy and efficiency.

Core Features & Use Cases

• Model Selection Guidance: Provides a comparative overview of various embedding models, their dimensions, token limits, and best use cases (e.g., code, finance, general text).
• Embedding Pipeline: Illustrates the process of document embedding, from chunking to vector generation.
• Code & Domain Specifics: Offers specialized templates for different embedding needs, including local development, code embedding, and domain-specific pipelines.
• Quality Evaluation: Includes functions to evaluate the performance of embedding models for retrieval tasks.
• Use Case: When building a RAG system for a legal document database, you can use this Skill to compare models like voyage-law-2 against general-purpose models to find the one that best captures the nuances of legal text.

Quick Start

Use the selecting-embeddings skill to get embeddings for the text 'This is a sample document.' using the text-embedding-004 model.