neo4j-graphrag-python and kg-rag
The VectorInstitute/kg-rag project likely serves as an **implementation example or research project** that demonstrates a comprehensive KG-RAG framework, while neo4j/neo4j-graphrag-python is a **production-ready library provided by Neo4j** specifically for building GraphRAG applications, implying that kg-rag could potentially *use or inspire* the techniques employed by the more mature Neo4j offering.
About neo4j-graphrag-python
neo4j/neo4j-graphrag-python
Neo4j GraphRAG for Python
Supports automated knowledge graph construction from unstructured text and PDFs via LLM-powered entity/relation extraction, alongside multiple retrieval strategies (vector search, graph traversal, hybrid, and Text2Cypher). Integrates with major LLM providers (OpenAI, Anthropic, Google, Cohere, Ollama, MistralAI) and optional external vector stores (Weaviate, Pinecone, Qdrant), with experimental NLP components using spaCy for semantic resolution.
About kg-rag
VectorInstitute/kg-rag
This project implements a comprehensive framework for Knowledge Graph Retrieval Augmented Generation (KG-RAG). It focuses on financial data from SEC 10-Q filings and explores how knowledge graphs can improve information retrieval and question answering compared to baseline approaches.
Implements multiple retrieval strategies including entity-based embedding matching with beam search, Cypher queries against Neo4j, and hierarchical community detection (GraphRAG-style), enabling direct comparison of knowledge graph approaches versus traditional vector similarity and chain-of-thought baselines. Built as a modular Python package with Chroma vector stores, OpenAI LLM integration, and comprehensive evaluation pipelines including hyperparameter search across methods.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work