Neura Market
Neo4j LLM Knowledge Graph Builder logo

Neo4j LLM Knowledge Graph Builder

Free

Neo4j graph construction from unstructured data using LLMs

FreeFree tier
Type
Open Source
Company
Neo4j
Visit Website GitHub (38 stars)

About Neo4j LLM Knowledge Graph Builder

The Neo4j LLM Knowledge Graph Builder is an open-source application that transforms unstructured data from various sources—such as PDFs, Word documents, text files, YouTube videos, web pages, and cloud storage (GCS, S3)—into structured knowledge graphs stored in Neo4j. It leverages Large Language Models (LLMs) and the LangChain framework to extract nodes, relationships, and properties. Users can select from a wide range of LLM providers (e.g., OpenAI, Gemini, Anthropic, Ollama, Groq), configure custom or existing graph schemas, visualize graphs in Neo4j Bloom, interact with data through conversational queries, and track token usage. The tool requires Python 3.12+, Neo4j 5.23+ with APOC, and supports Neo4j Aura (including free tier).

Key Features

Transforms unstructured data (PDFs, DOCs, TXTs, YouTube videos, web pages) into structured knowledge graphs
Supports multiple LLM providers: OpenAI, Gemini, Diffbot, Azure OpenAI, Anthropic, Fireworks, Groq, Amazon Bedrock, Ollama, Deepseek, and OpenAI-compatible base URL models
Custom schema support or use of existing schemas for graph generation
Graph visualization in Neo4j Bloom for single or multiple data sources
Chat with data: conversational queries against the Neo4j database with metadata about response sources
Token usage tracking per user and database connection, with daily/monthly limits and API monitoring
Embedding model selection configurable from the frontend for vector embeddings
File upload from local machine, Google Cloud Storage (GCS), S3 bucket, or web sources

Pros & Cons

Pros
  • Free and open-source with a permissive license (Apache 2.0 implied by GitHub)
  • Supports a wide variety of LLM providers and embedding models for flexibility
  • Integrates seamlessly with Neo4j ecosystem including Bloom visualization and Aura cloud databases
  • Handles multiple data source types (local, cloud storage, web) from a single interface
  • Token usage tracking helps manage LLM costs and limits
Cons
  • Requires Neo4j database 5.23+ with APOC installed, adding setup overhead
  • Neo4j Desktop users must deploy backend and frontend separately (no docker-compose support)
  • Heavy reliance on external LLM APIs may incur costs despite the tool being free
  • Setup requires Python 3.12+ and environment configuration, which may be complex for non-technical users

Best For

Converting uploaded PDFs and documents into knowledge graphs for semantic search and analysisExtracting structured relationships from YouTube video contentBuilding knowledge graphs from web pages and online documentationEnabling interactive Q&A over graph data with conversational chat interfaceGenerating graphs based on custom schemas for domain-specific applications

FAQ

What data sources are supported?
You can upload files from local machine, Google Cloud Storage (GCS), S3 bucket, or web sources. Supported file types include PDFs, DOCs, TXTs, YouTube videos, and web pages.
Which LLMs can be used?
The tool supports OpenAI, Gemini, Diffbot, Azure OpenAI, Anthropic, Fireworks, Groq, Amazon Bedrock, Ollama, Deepseek, and any OpenAI-compatible base URL models.
What are the system requirements?
Python 3.12 or higher is required. The backend uses Neo4j 5.23 or later with APOC installed. Neo4j Aura (including free tier) is supported. For Neo4j Desktop, separate deployment of backend and frontend is required.
Can I visualize the generated knowledge graph?
Yes, you can visualize graphs for specific or multiple data sources simultaneously using Neo4j Bloom.
Is there a free tier?
The tool itself is open-source and free. You only need a Neo4j instance (Aura offers a free tier) and optionally pay for LLM API usage.