GraphAide

Build knowledge graphs from any document with LLMs and Neo4j. GraphAide is a multi-agentic platform for building intelligent applications powered by Large Language Models (LLMs). Knowledge graph construction is one powerful example—transforming documents into queryable knowledge graphs by extracting entities and relationships. But GraphAide's capabilities extend far beyond KGs.

Quick Start

Install (Version 0.6.4)

Using uv (Recommended - Fastest):

uv pip install graphaide==0.6.4

Or using pip:

pip install graphaide==0.6.4

30-Second Example

from graphaide import GraphAide, ModelConfig, get_sample_data_path

# Setup
graphaide = GraphAide(
    model_config=ModelConfig(
        provider="openai", 
        model_name="gpt-4o",
        api_key="..."
    )
)

# Extract from sample document
sample_pdf = get_sample_data_path("PNNL_About.pdf")
result = graphaide.extract(file_path=sample_pdf)
print(f"✓ Extracted {len(result.nodes)} entities, {len(result.edges)} relationships")

What It Does

• Extract entities & relationships from PDFs, text, images, or web data
• Process diverse data with multiple LLM providers and models
• Ground information to external knowledge bases and sources
• Build intelligent workflows with minimal setup
• Track data provenance and attribution throughout
• Scale from single documents to large datasets

Documentation

📚 Full documentation: https://pnnl.github.io/GraphAide

Getting Started:

• Installation — Setup instructions for uv, pip, Docker, and development
• Configuration — Configure LLMs, Neo4j, vector stores
• Extraction Guide — Learn entity and relationship extraction
• Vector Stores — Setup ChromaDB for semantic search
• Ingestion — Load extracted data into Neo4j
• Resources — Tutorials and examples

Supported Providers

Use Cases

• Knowledge Management — Build searchable knowledge graphs from company documents
• Research Analysis — Extract and connect findings across papers and reports
• Data Integration — Merge information from multiple sources with automatic entity linking
• Compliance Auditing — Track relationships and dependencies with full provenance
• Domain Expertise — Answer questions requiring reasoning across documents

Project Status

• Version: 0.6.4
• Status: Active Development
• License: MIT

Citation

If you use GraphAide in your research, please cite:

@inproceedings{purohit2024graphaide,
  title={GraphAide: Advanced Graph-Assisted Query and Reasoning System},
  author={Purohit, Sumit and Chin, George and Mackey, Patrick S and Cottam, Joseph A},
  booktitle={2024 IEEE International Conference on Big Data (BigData)},
  pages={3485--3493},
  year={2024},
  organization={IEEE}
}

Support & Contributing

• 📚 See full documentation for detailed guides
• 🐛 Report issues on GitHub Issues
• 💬 Ask questions via GitHub Discussions
• 📖 Check QUICKSTART.md for detailed setup

Legal

• License — MIT License
• Disclaimer — Important legal notices
• Acknowledgments — Credits and funding

Ready to build your first knowledge graph? Start with Installation.