AllegroGraph 9.0 Launches with GraphTalker, the AI Agent for Enterprise Knowledge Graphs

GraphTalker enables business and technical users to ask natural-language questions of their Knowledge Graphs and receive verified answers, data analysis, and reusable query workflows. Unlike traditional natural-language-to-query systems that attempt a one-shot translation from prompt to query, GraphTalker uses an iterative, agent-driven approach that explores schema, tests queries, learns from errors, and refines results until it produces a working answer.

With GraphTalker, AllegroGraph v9.0 advances Franz’s vision for enterprise AI where Knowledge Graphs are the semantic layer for agent-ready data, combining symbolic reasoning, graph analytics, vector search, and LLM-powered natural-language interaction.

“GraphTalker represents a major step forward for enterprise AI because it moves beyond simple chat and into agentic interaction with the Knowledge Graph itself,” said Dr. Jans Aasman, CEO of Franz Inc. “It can inspect the schema, understand relationships, test hypotheses, refine queries, and explain its reasoning. That is exactly the kind of Neuro-Symbolic foundation enterprises need for trusted, explainable AI agents.”

New Capabilities in AllegroGraph v9.0 include:

GraphTalker AI Agent: A natural-language interface that allows users to query, analyze, and reason over enterprise Knowledge Graphs without needing to write SPARQL.

Iterative Query Generation and Validation: GraphTalker does not simply generate a query and hope for the best. It explores the semantic layer, examines schema, reviews examples, tests queries, interprets failures, and improves its approach step by step.

Schema-Aware Intelligence: GraphTalker uses SHACL-based shape descriptions and navigable schema overviews so the agent can understand classes, properties, and relationships before generating queries.

Stateful, Multi-Turn Analysis: Users can refine questions, pursue follow-ups, and maintain context across longer analytical sessions.

Reusable Knowledge Workflows: GraphTalker can turn answered questions into reusable, schedulable, and governed data workflows, making insights discovered through natural language durable enterprise processes.

Embed GraphTalker as an Independent Agentinto End-User Applications: GraphTalker can be integrated directly into enterprise applications, portals, dashboards, notebooks, and AI agent workflows, allowing organizations to deliver natural-language Knowledge Graph interaction to their own users. Through its HTTP-based API and evaluation server, developers can submit natural-language questions from any application and return structured, explainable answers powered by AllegroGraph.

Enterprise Semantic Layer for Agentic AI: AllegroGraph v9.0 strengthens the role of the Knowledge Graph as the governed, explainable, and standards-based semantic layer that connects enterprise data to AI agents.

GraphTalker has been applied across enterprise HR analytics, product catalogs, biomedical Knowledge Graphs, and geospatial data. Without retraining or domain-specific tuning, it adapts by exploring schema and data at query time, giving organizations a practical way to scale natural-language access across multiple domains.

Read more in the Technical Documentation.

Get Started Today.