Native Neo4j integrations with Google Cloud will accelerate gen AI action

Neo4j, the graph database and analytics company, has announced new native integrations with Google Cloud that speed up generative AI (gen AI) application development and deployment. The results address complexity and hallucinations when building and deploying successful gen AI applications requiring real-time, contextually rich data and accurate, explainable results.

Knowledge graphs support gen AI. They capture relationships between entities, ground LLMs in facts, and enable LLMs to reason, infer, and retrieve relevant information accurately and effectively.

According to Gartner’s November 2023 report AI Design Patterns for Knowledge Graphs and Generative AI, “data and analytics leaders must leverage the power of large language models (LLMs) with the robustness of knowledge graphs for fault-tolerant AI applications.”

Retrieval Augmented Generation (RAG) is the technique by which LLMs access external datasets. Combining knowledge graphs with RAG, known as GraphRAG, ensures that gen AI outcomes are accurate, explainable, and transparent, including with real-time data.

Developers can easily apply GraphRAG techniques with knowledge graphs to ground LLMs for accuracy, context, and explainability, enhancing gen AI innovation. Specifically, they can:

Quickly create knowledge graphs for accurate, explainable results

Developers can easily create knowledge graphs with Gemini models, Google Cloud VertexAI, LangChain, and Neo4j from unstructured data like PDFs, web pages, and documents – either directly or loaded from Google Cloud Storage buckets.

Ingest, process, and analyse real-time data in seconds

Developers can use Flex templates in Dataflow to create repeatable, secure data pipelines that ingest, process, and analyse data across Google BigQuery, Google Cloud Storage, and Neo4j—supplying knowledge graphs with real-time information and enabling gen AI applications to provide relevant, timely insights.

Build gen AI applications powered by knowledge graphs on Google Cloud

Customers can use Gemini for Google Workspace and Reasoning Engine from the Vertex AI platform to easily deploy, monitor, and scale gen AI apps and APIs onto Google Cloud Run. Gemini models are trained on Neo4j’s training data to automatically turn any language code snippets to Neo4j’s Cypher query language. The result makes application development faster, easier, and more collaborative by integrating natural language understanding and generation capabilities within various applications and environments.

Developers can also use Cypher with any integrated development environment (IDE) supported by Gemini models for more efficient querying and visualisation of graph data. Neo4j’s vector search, GraphRAG, and conversational memory capabilities integrate seamlessly through LangChain and Neo4j AuraDB with Google Cloud.

Regulated customers will be able to meet strict data residency, security, and regulatory requirements in Google Distribution Cloud (GDC) Hosted, which became generally available in March. GDC is an air-gapped private cloud infrastructure and edge environment designed specifically for public sector organisations and regulated enterprises. Neo4j is the preferred launch partner for GDC to provide Graph Database and Analytics capabilities. 

Customers can also run in-memory graph analysis of complex hidden data patterns using Neo4j’s catalogue of 70+ graph data science functions directly on BigQuery data and from BigQuery SQL using Apache Spark Stored Procedures.

“Generative AI can significantly increase the value customers get from critical business data. By utilising Google Cloud’s Gemini models and Vertex AI, Neo4j can increase the speed and accuracy of generative AI application development,” said Ritika Suri, Director of Technology Partnerships, Google Cloud.

Added Sudhir Hasbe, Chief Product Officer, Neo4j: “GraphRAG with Neo4j and Google Cloud enables enterprises to move from gen AI development to deployment much faster and see value from their production use cases. Our latest milestone combines the power of graph technology, gen AI, and cloud computing excellence, enabling enterprises to achieve better results faster from their connected data, and innovate with gen AI.”

Neo4j also integrated native vector capabilities into its core graph database last year, enabling it to serve as long-term memory for LLMs.

Said Jeff Dalgliesh, CTO, Data², which builds intelligence and reasoning engineering assistants for the intelligence and oil and gas sectors: "Data² uses generative AI and knowledge graphs to enable organisations to maximise the potential of their data. (The) announcement of Neo4j’s new native integrations with Google Cloud, particularly the ability to enhance gen AI applications with Neo4j knowledge graphs, marks an exciting step forward for the industry.

"With Neo4j’s GraphRAG approach and Google Cloud’s robust infrastructure, we will be able to deliver even more powerful, explainable AI insights to our clients, helping them make critical decisions with confidence and speed." 

Moheesh Raj, Senior Engineering Manager, Dun & Bradstreet said: "At Dun & Bradstreet, we’re committed to helping companies leverage data and analytical insights to take more intelligent actions that deliver a competitive edge. (The) announcement opens up exciting new possibilities for us to further streamline and accelerate our compliance services, setting a new standard for data-driven due diligence in the industry."

These capabilities are available now.