microsoft/graphrag

A modular graph-based Retrieval-Augmented Generation (RAG) system

What it solves

GraphRAG leverages knowledge graphs to help LLMs reason better about private, unstructured narrative data. It addresses the problem of extracting structured meaning from complex text to provide a more targeted context for question answering.

How it works

It functions as a data pipeline and transformation suite. It uses LLMs to extract structured data from unstructured text, creating a knowledge graph memory structure that the LLM can then use to enhance its outputs.

Who it’s for

Developers and researchers who need to enhance an LLM's ability to perform discovery and reasoning over their own private datasets.

Highlights

  • own private data
  • knowledge graph memory structures
  • data pipeline and transformation suite
  • prompt tuning capabilities

Related

  • Project
    rahulnyk/knowledge_graph

    A tool that converts text corpora into interactive knowledge graphs of concepts using local LLMs, enabling advanced text analysis and Graph RAG.

  • Project
    InternScience/GraphGen

    A framework for generating knowledge-driven synthetic data using knowledge graphs to fill knowledge gaps in LLMs for improved supervised fine-tuning.

  • Project
    HKUDS/LightRAG

    A lightweight, graph-based RAG framework that combines knowledge graphs and vector embeddings to provide efficient, high-quality retrieval and global context understanding.

  • Project
    LHRLAB/HyperGraphRAG

    A Retrieval-Augmented Generation framework that uses hypergraph-structured knowledge representation to better capture complex relationships between entities.

  • Project
    DEEP-PolyU/LinearRAG

    LinearRAG is a relation-free graph construction method for GraphRAG that eliminates LLM token costs during graph building to enable efficient, scalable multi-hop reasoning on large corpora.