AI product Open source
An open-source Python package from Neo4j for building graph retrieval-augmented generation applications with Neo4j. It provides knowledge-graph construction pipelines that process text or PDFs, use an LLM to extract schema-constrained entities and relationships, generate embeddings, and populate a Neo4j database. The package includes vector-index utilities, embedding and LLM integrations, GraphRAG generation, and retrievers for vector, hybrid, graph-traversal, and text-to-Cypher retrieval. Its `Pipeline` supports extensive customization, while `SimpleKGPipeline` provides a simpler abstraction; optional dependencies add support for providers and external vector stores including Weaviate, Pinecone, and Qdrant. It is installable with pip or conda and supports Python 3.10 through 3.14, with experimental knowledge-graph features identified as unsuitable for production use.
1 use taken from transcripts — each links to the moment in the video.
An open-source Python package from Neo4j for knowledge graph building and retrieval and agentic work. The workshop uses it in a GitHub Codespace to run a knowledge graph builder that chunks documents, extracts entities and relationships with an LLM, generates embeddings, and populates a Neo4j database.
1 in the library.