LLM Graph Builder

A no-code application that uses LLMs to transform unstructured PDFs, web pages, and transcripts into structured Neo4j knowledge graphs.

Neo4j LLM Graph Builder streamlines the creation of knowledge graphs by automating entity and relationship extraction from raw text. Built on the LangChain framework (specifically the llm-graph-transformer module), it supports models like GPT-4o, Gemini 1.5 Pro, and Claude 3.5 Sonnet to process diverse inputs including YouTube transcripts and S3 buckets. The tool maps data into a dual-layer structure: a lexical graph for document chunks and an entity graph for semantic connections. This architecture enables GraphRAG (Graph Retrieval-Augmented Generation), allowing users to query their data through a conversational interface that provides explainable, multi-hop insights directly from a Neo4j Aura database.

https://llm-graph-builder.neo4jlabs.com