AI/ML Optimization Framework
4.7k 2026-04-18
Marker-Inc-Korea/AutoRAG
An open-source framework that automates the evaluation and optimization of Retrieval-Augmented Generation (RAG) pipelines using AutoML-style techniques for specific datasets.
Core Features
Automated RAG pipeline optimization for custom data.
Comprehensive evaluation of various RAG modules.
Tools for creating QA and Corpus datasets for RAG.
Support for custom Large Language Models (LLMs) and embedding models.
Interactive dashboard for RAG optimization results.
Quick Start
pip install AutoRAGDetailed Introduction
AutoRAG addresses the challenge of identifying the most effective RAG pipeline for unique datasets and use-cases. Given the multitude of RAG modules and configurations, manual evaluation is time-consuming and inefficient. AutoRAG provides an AutoML-style automation framework to systematically evaluate and optimize different RAG module combinations, helping users automatically discover the best-performing RAG pipeline tailored to their specific needs, thereby enhancing RAG performance and development efficiency.