Franz Inc., an early innovator in AI and leading supplier of Graph Database technology for Neuro-Symbolic AI solutions, is launching AllegroGraph 9.1—featuring GraphTalker 2.0, a major advancement of its agentic natural-language interface for enterprise Knowledge Graphs.
Building on the introduction of GraphTalker in AllegroGraph 9.0, the new release enables organizations to more effectively manage the contextual knowledge that guides AI interactions, incorporate documents and images into graph-based analysis, and preserve reusable knowledge and analytical workflows in a persistent, learning enterprise memory layer, according to Franz.
Through collaborative learning between the context graph and human domain experts, this persistent memory retains feedback, corrections, and insights that inform future analysis.
At the core of GraphTalker 2.0 is a new Context Management System that allows organizations to create, edit, version, and reuse the domain-specific knowledge that helps AI agents understand their enterprise data. This includes ontologies, relationships, business terminology, domain conventions, and proven query patterns, the company said.
"AI agents need more than access to data,” said Dr. Jans Aasman, CEO of Franz Inc. “They need to understand what that data means, how it relates to other information, and the business context in which decisions are made. With GraphTalker 2.0, we're making that context easier to manage, preserve, and reuse—and enable human domain experts to enrich it through ongoing feedback and collaboration. This is an important step toward the Context Graph as a persistent enterprise memory layer, where AI agents can combine structured knowledge, business rules, and historical context to deliver more reliable and explainable results."
New Capabilities in AllegroGraph 9.1 include:
- GraphTalker 2.0: A major upgrade of AllegroGraph's agentic natural-language interface, featuring an entirely redesigned user experience and improved management of graph-based analytical workflows.
- Enterprise Context Management: Create, edit, upload, and manage repository-specific context files that guide GraphTalker's understanding of ontologies, business terminology, data relationships, and query conventions.
- Context Versioning and Portability: Track modifications, compare file versions, restore previous content, and export or import context files with their version histories across repositories.
- Document and Image Attachments: Incorporate PDFs, images, text documents, and CSV files into natural-language requests, enabling analysis that combines external information with enterprise Knowledge Graph data.
- Streamlined User Interface: Navigate between Chat, Context, Queries, and History through a unified interface, with simplified switching between repositories and centralized settings.
- Persistent Queries, Visualizations, and Sessions: Save and revisit generated queries, visualizations, and analytical sessions through a reorganized storage architecture designed to support continued knowledge exploration.
- Enhanced Support for Large Knowledge Graph Schemas: Preprocessed schema overviews help GraphTalker work with complex enterprise data models containing large numbers of classes, properties, and relationships.
For more information about this news, visit https://franz.com.