Skip to content

iText2KG

Build knowledge graphs from text with LLMs

Visit official site

What iText2KG is

iText2KG is described as a Python package that uses large language models to extract entities and relations and build a knowledge graph step by step.

What you can do with it

  • Distil documents into semantic blocks
  • Extract unique entities and their relations
  • Visualise the graph in Neo4j

Getting started

  • Install iText2KG with pip in Python
  • Define or upload a distillation schema
  • Use DocumentDistiller to get semantic blocks
  • Build the graph with the iText2KG module

Pros & cons

✓ What we like

  • Zero-shot extraction needs no domain training
  • Entities and relations are resolved for uniqueness

! What to watch out for

  • Graph output is tied to Neo4j
  • Building starts from a defined distillation schema

FAQ

What does it extract?

The material names entities and relations from text documents.

How is the graph shown?

The material describes visualisation through Neo4j.

Last reviewed: 2026-11-15

More AI coding tools tools

View all →

How we review