microsoft/graphrag
A modular graph-based Retrieval-Augmented Generation (RAG) system
This system helps you make sense of large amounts of unstructured text data, like research papers or internal documents. It processes your text to identify key entities and relationships, outputting a structured knowledge graph that your AI can then use to answer complex questions or find insights more effectively. This is designed for researchers, analysts, or anyone who needs to extract precise information and reasoning from extensive narrative data using large language models.
31,429 stars. Used by 1 other package. Actively maintained with 8 commits in the last 30 days. Available on PyPI.
Use this if you need to boost your AI's ability to reason about complex, private narrative data by transforming it into a structured knowledge graph.
Not ideal if you have very small datasets or primarily structured data, or if you are not prepared for potentially high processing costs during initial setup.
Stars
31,429
Forks
3,319
Language
Python
License
MIT
Category
Last pushed
Mar 12, 2026
Commits (30d)
8
Dependencies
25
Reverse dependents
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/microsoft/graphrag"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Community Discussion
Recent Releases
Compare
Related tools
neo4j/neo4j-graphrag-python
Neo4j GraphRAG for Python
Hawksight-AI/semantica
Semantica 🧠 — A framework for building semantic layers, context graphs, and decision...
FalkorDB/GraphRAG-SDK
Build fast and accurate GenAI apps with GraphRAG SDK at scale.
getzep/graphiti
Build Real-Time Knowledge Graphs for AI Agents
gusye1234/nano-graphrag
A simple, easy-to-hack GraphRAG implementation