lizhe2004/Awesome-LLM-RAG-Application

What it solves

This project provides a comprehensive, curated collection of resources for building Retrieval-Augmented Generation (RAG) applications. It helps developers and researchers navigate the rapidly evolving landscape of RAG by organizing academic papers, open-source tools, and industry best practices into a single, structured reference point.

How it works

The repository acts as a knowledge hub that categorizes RAG components into logical stages, such as data preprocessing, retrieval strategies, evaluation frameworks, and LLM deployment. It includes links to foundational research papers, technical guides, and a wide array of open-source frameworks (like LangChain, LlamaIndex, and Dify) that developers can use to build their own RAG systems.

Who it’s for

This list is designed for AI engineers, researchers, and developers who are building or studying LLM-based applications. It is particularly useful for those looking to move from initial prototypes to production-ready enterprise systems, as it covers everything from basic concepts to advanced techniques like GraphRAG and agentic workflows.

Highlights

• Academic Research: A chronological list of key papers covering RAG evolution, evaluation, and agentic search.
• Tooling Ecosystem: A categorized directory of open-source frameworks for RAG, document preprocessing, and observability.
• Practical Guidance: Links to enterprise-level practices, tutorials, and comparisons between RAG and fine-tuning.
• Advanced Techniques: Resources on query optimization, re-ranking, and context compression to improve retrieval accuracy.
• Multi-modal Support: References for handling diverse data types including images, tables, and PDFs.