RAG_Techniques and RAG-ARC

These are complements: the first is an educational resource demonstrating RAG techniques and patterns that could inform the architectural decisions implemented in the second's modular framework.

RAG_Techniques
57
Established
RAG-ARC
50
Established
Maintenance 10/25
Adoption 10/25
Maturity 16/25
Community 21/25
Maintenance 10/25
Adoption 7/25
Maturity 15/25
Community 18/25
Stars: 25,887
Forks: 3,041
Downloads:
Commits (30d): 0
Language: Jupyter Notebook
License:
Stars: 38
Forks: 13
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No Package No Dependents
No Package No Dependents

About RAG_Techniques

NirDiamant/RAG_Techniques

This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. RAG systems combine information retrieval with generative models to provide accurate and contextually rich responses.

Covers advanced RAG patterns including agentic retrieval loops, hybrid search strategies (dense-sparse retrieval fusion), query optimization techniques, and multi-document reasoning—beyond basic retrieval pipelines. Implementations target popular frameworks like LangChain and LlamaIndex with code-first Jupyter notebooks, focusing on practical enhancements for production-grade systems.

About RAG-ARC

DataArcTech/RAG-ARC

A modular, high-performance Retrieval-Augmented Generation framework with multi-path retrieval, graph extraction, and fusion ranking

Supports multi-format document parsing (PDF, DOCX, PPT, Excel) with OCR and layout-aware strategies, combining sparse (BM25), dense (FAISS-GPU), and full-text search via Reciprocal Rank Fusion. Built on FastAPI with PostgreSQL, Redis, and Neo4j integration, enabling incremental knowledge graph updates and GraphRAG with subgraph PPR for efficient reasoning.

Scores updated daily from GitHub, PyPI, and npm data. How scores work