This GitHub organization collects relevant projects and resources like the pattern catalog (repository) and more.

Retrieval Augmented Generation (RAG) is a way to ground Large Language Models (LLMs) in factual data to reduce hallucinations and extend the information available for question answering. The user's question will be used to retrieve relevant information from one or more data sources. These facts provide a ground source of truth that is used to augment the context of a prompt. The augmented prompt and original user question are then passed to the LLM to generate the answer.

GraphRAG are retrieval mechanisms that use graph structures to provide more fine-grained and relevant contextual information than a plain text search (or vector search) would. As it can utilize the rich representation of knowledge about many areas in a knowledge graph.

Contributions are very welcome.

A collection of resources around GraphRAG a set of advanced GenAI RAG (Retrieval Augmented Generation) patterns.