llm-graph-builder

Neo4j graph construction from unstructured data using LLMs

What is llm-graph-builder?

Neo4j graph construction from unstructured data using LLMs

Overview of llm-graph-builder

What it is

llm-graph-builder is an experimental Neo4j Labs project that transforms unstructured data (PDFs, DOCs, TXTs, YouTube videos, web pages, etc.) into a structured Knowledge Graph stored in Neo4j. It uses Large Language Models (LLMs) and the LangChain framework. The application allows users to upload files from local machines, GCS, S3 buckets, or web sources, choose an LLM model, and generate a Knowledge Graph.

Key features

  • Knowledge Graph Creation: Transforms unstructured data into structured Knowledge Graphs using LLMs, extracting nodes, relationships, and their properties.
  • Schema Support: Users can use a custom schema or existing schemas configured in settings to generate graphs.
  • Graph Visualization: View graphs for specific or multiple data sources simultaneously in Neo4j Bloom.
  • Chat with Data: Interact with data in the Neo4j database through conversational queries; retrieve metadata about the source of responses.
  • LLMs Supported: Includes OpenAI, Gemini, Diffbot, Azure OpenAI, Anthropic, Fireworks, Groq, Amazon Bedrock, Ollama, Deepseek, and other OpenAI-compatible base URL models.
  • Token Usage Tracking: Monitor LLM token usage per user and database connection; view daily and monthly token consumption and limits.
  • Embedding Model Selection: Choose from embedding models (OpenAI, Gemini, Amazon Titan, Sentence Transformers) to generate vector embeddings.

Use cases

llm-graph-builder is suitable for converting unstructured documents (PDFs, DOCs, TXTs), YouTube videos, web pages, and other text sources into structured knowledge graphs. It can be used for data integration, knowledge management, and building graph-based applications that require entity and relationship extraction from diverse sources.

Getting started

Prerequisites: Python 3.12 or higher, Neo4j Database 5.23 or later with APOC installed. Neo4j Aura databases (including free tier) are supported. For Neo4j Desktop, deploy backend and frontend separately (docker-compose not supported).Backend setup:bashcd backendpython3.12 -m venv venvsource venv/bin/activate # On Windows: venv/Scripts/activatepip install -r requirements.txtuvicorn score:app --reloadFrontend setup:bashcd frontendyarnyarn run devFor Docker-Compose deployment, see the repository README.

License and pricing

License: Apache-2.0. Pricing: open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/neo4j-labsllm-graph-builder.svg)](https://yunodb.com/tools/neo4j-labsllm-graph-builder?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/neo4j-labsllm-graph-builder?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/neo4j-labsllm-graph-builder.svg" alt="Featured on YunoDB" height="20"></a>