Image of an entity network with color-coded communities

Project GraphRAG

LLM-Derived Knowledge Graphs

GraphRAG

GraphRAG (Graphs + Retrieval Augmented Generation) is a technique for richly understanding text datasets by combining text extraction, network analysis, and LLM prompting and summarization into a single end-to-end system.

Read the pre-print
GraphRAG image on blue to green gradient

GraphRAG auto-tuning provides rapid adaptation to new domains 

September 9, 2024

GraphRAG blog hero

GraphRAG: New tool for complex data discovery now on GitHub 

July 2, 2024 | Darren Edge, Ha Trinh, Steven Truitt, Jonathan Larson

Project Ire - GraphRag background: Blue-green gradient

GraphRAG: Unlocking LLM discovery on narrative private data 

February 13, 2024 | Jonathan Larson, Steven Truitt

Three icons that represent (left to right) search, GraphRAG, and the globe.

Introducing DRIFT Search: Combining global and local search methods to improve quality and efficiency 

October 31, 2024

The image features three white icons on a gradient background transitioning from blue on the left to green on the right. The first icon, located on the left, depicts a hierarchical structure resembling a workflow with connected squares. The middle icon represents GraphRAG (interconnected nodes and lines). The third icon, on the right, shows a globe with a magnifying glass overlaying it, symbolizing global search.

GraphRAG: Improving global search via dynamic community selection 

November 15, 2024 | Bryan Li, Ha Trinh, Darren Edge, Jonathan Larson

Open Source

The GraphRAG library and complementary solution accelerator, an easy-to-use API experience hosted on Azure, are now both available on GitHub!

GraphRAG Solution Accelerator (opens in new tab)

GraphRAG Library (opens in new tab)