Starting today, you can build Graph Retrieval-Augmented Generation (GraphRAG) applications by combining knowledge graphs stored in Amazon Neptune and LlamaIndex, a popular open-source framework for building applications with Large Language Models (LLM) such as those available in Amazon Bedrock.
Customers looking to build Generative AI applications often use Retrieval-Augmented Generation (RAG) to improve LLM output so it remains relevant, accurate, and useful in various contexts. RAG extends the already powerful capabilities of LLMs to specific domains or an organization’s internal knowledge base, without the need to retrain the model. Knowledge graphs explicitly consolidate and integrate an organization’s information assets. GraphRAG uses knowledge graphs, existing graphs or ones generated from source data, to relate concepts and entities across the underlying content, further improving RAG applications. For example, if asked, “Tell me about news events that impact companies in my trading portfolio,” a GraphRAG app could respond by also identifying news articles for upstream and downstream dependencies in the supply chain, which in turn might have an impact on those companies. With today’s announcement , you can use LlamaIndex to create GraphRAG applications with knowledge graphs stored in Amazon Neptune.
To get started visit the Amazon Neptune GraphStore documentation.
Source:: Amazon AWS