GraphRAG
Graph-based retrieval-augmented generation from Microsoft Research
GraphRAG is profiled here as a RAG Framework tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
GraphRAG is an open-source approach from Microsoft Research that builds a knowledge graph from a document collection and uses it to answer questions that span many sources. Where ordinary retrieval pulls isolated text chunks, GraphRAG extracts entities and relationships with a language model, clusters them into communities, and generates summaries that let the system reason over the whole corpus. This structure improves answers to broad, thematic questions where the relevant facts are scattered across many documents.
Key Capabilities:
Knowledge-graph construction that extracts entities and relationships from text
Community detection that groups related entities for hierarchical summarization
Global search that answers thematic questions across an entire corpus
Local search that combines graph context with relevant text chunks
A configurable indexing pipeline for building graphs from raw documents
Provenance that links generated answers to source entities
Alternative tools
- RAGFlow
Open-source RAG engine with deep document understanding
- Milvus
Open-source vector database built for billion-scale search
- txtai
All-in-one embeddings database for semantic search and RAG
- R2R
Production retrieval system with ingestion and an API
- Chroma
Developer-first embedding database that runs anywhere
- Qdrant
Rust-based vector search engine with rich filtering