Skip to content

Docling Graph Documentation

Docling Graph

What is Docling Graph?

Docling-Graph turns documents into validated Pydantic objects, then builds a directed knowledge graph with explicit semantic relationships.

This transformation enables high-precision use cases in chemistry, finance, and legal domains, where AI must capture exact entity connections (compounds and reactions, instruments and dependencies, properties and measurements) rather than rely on approximate text embeddings.

This toolkit supports two extraction paths: local VLM extraction via Docling, and LLM-based extraction using either local runtimes (vLLM, Ollama, LM Studio) or API providers (Mistral, OpenAI, Gemini, IBM watsonx, Amazon Bedrock), all orchestrated through a flexible, config-driven pipeline.

Key Features

  • ✍🏻 Multi-Format Input: PDF, images, Office, HTML, Markdown, URLs, and more.
  • 🧠 Flexible Extraction: LLM/VLM backends with configurable pipelines.
  • ⚙️ Type-Safe IR: Pydantic-based information retrieval.
  • 💎 Knowledge Graphs: Pydantic to NetworkX conversion with stable IDs.
  • 📍 Provenance: Deterministic source grounding with no extra LLM calls.
  • 📦 Multiple Exports: CSV, Cypher, JSON, Markdown, and Neo4j-ready formats.
  • 📊 Visual Reports: Interactive HTML and Markdown.

Quick Navigation

Getting Started

Core Documentation

  • Introduction

    Overview, architecture, and core concepts

  • Fundamentals

    Installation, schema definition, pipeline configuration, extraction, and more

  • Usage

    CLI reference, Python API, examples, and advanced topics

  • Reference

    Detailed API documentation

  • Community

    Contributing and development guide

Resources

Documentation

Community

Next Steps

  1. Install Docling Graph
  2. Follow the Quick Start
  3. Create Your First Template
  4. Explore Examples

Need Help?