Franz adds GraphTalker to AllegroGraph 9.0 for enterprise knowledge graphs

By AI, Created 10:50 AM UTC, May 20, 2026, /AGP/ – Franz Inc. launched AllegroGraph 9.0 on May 5, 2026, adding GraphTalker, an AI agent that lets enterprise users query and reason over knowledge graphs in natural language. The release pushes the company’s neuro-symbolic AI pitch by combining graph reasoning, vector search and LLM interaction in a governed semantic layer.

Why it matters: - AllegroGraph 9.0 is aimed at making enterprise knowledge graphs usable by business and technical teams without requiring SPARQL. - GraphTalker is designed to return verified answers, analysis and reusable workflows, which could turn one-off questions into governed enterprise processes. - Franz is positioning the knowledge graph as the semantic layer for agent-ready data in neuro-symbolic AI systems.

What happened: - Franz Inc. announced AllegroGraph v9.0 on May 5, 2026, with GraphTalker as a new AI agent for enterprise knowledge graphs. - GraphTalker gives users a natural-language way to explore, query and reason over knowledge graphs. - The release was tied to Franz’s broader enterprise AI strategy built around neuro-symbolic AI.

The details: - GraphTalker uses an iterative, agent-driven workflow instead of a one-shot natural-language-to-query translation. - The agent explores schema, tests queries, learns from errors and refines results until it produces a working answer. - GraphTalker uses SHACL-based shape descriptions and schema overviews to understand classes, properties and relationships before generating queries. - Users can maintain context across multi-turn analytical sessions and follow up on earlier questions. - GraphTalker can convert answered questions into reusable, schedulable and governed data workflows. - Developers can embed GraphTalker into enterprise applications, portals, dashboards, notebooks and AI agent workflows. - AllegroGraph’s HTTP-based API and evaluation server let applications submit natural-language questions and receive structured, explainable answers. - Franz says AllegroGraph v9.0 strengthens the role of the knowledge graph as a governed, explainable and standards-based semantic layer. - The company says GraphTalker has been applied across HR analytics, product catalogs, biomedical knowledge graphs and geospatial data. - Franz says GraphTalker adapts to new domains without retraining or domain-specific tuning by exploring schema and data at query time. - AllegroGraph combines symbolic reasoning, graph analytics, vector search and LLM-powered natural-language interaction.

Between the lines: - The release reflects a push beyond chat-style AI tools toward systems that can inspect structure, test hypotheses and explain reasoning. - That matters for enterprises that want AI output they can audit, govern and reuse. - Franz is also trying to differentiate AllegroGraph from standard natural-language query tools by emphasizing validation and iterative refinement. - Gartner listed Franz as a neuro-symbolic AI vendor in its 2025 Hype Cycle for AI, and Forrester said graphs are a foundational knowledge model for AI.

What’s next: - Dr. Jans Aasman is scheduled to join a panel at The Knowledge Graph Conference on May 5, 2026, titled “From Knowledge Graphs to Digital Twins: Are Our Models Ready?” - Franz will likely use the new release to push adoption across enterprise AI applications that need explainable graph-based reasoning. - The company’s broader market message is that knowledge graphs can serve as the durable data layer for agentic AI.

The bottom line: - AllegroGraph 9.0 is a product update and a strategic statement: Franz wants knowledge graphs to sit at the center of enterprise AI, not on the sidelines.