NVIDIA releases Auto Ontology for enterprise agent context
NVIDIA released Auto Ontology, an Apache 2.0-licensed Python framework that builds semantic, graph-based reasoning layers over enterprise data and metadata. Its repository says the framework can generate ontologies from metadata, reconcile inconsistent definitions, validate queries, and expose context with provenance through MCP, CLI, and API. The README also describes natural-language SQL querying powered by NVIDIA NIM, while benchmark instructions point to a separate repository without reporting performance results.
Auto Ontology offers an open-source way to give agents governed business definitions, validation, and provenance when working with enterprise data, although its claimed reliability and context-retrieval benefits have not been independently evaluated.
Auto Ontology is a Python framework for building semantic, graph-based reasoning layers over data. It enables deterministic query generation, better context retrieval, and more explainable AI for LLM and agent workflows.
Auto Ontology
Agents can retrieve documents and query databases, but reliability gates agentic analytics: a revenue metric answered faster is worth nothing if the number is wrong. Agents need a context layer that tells them where data lives, what it means, whether to use it, how to use it, where it came from, and what is similar. NVIDIA Auto Ontology is an open library that builds that layer from enterprise metadata, reconciles definitions across systems, and exposes it to any agent harness through MCP, CLI, and API. It complements the semantic models customers already own.
It enables developers to:
- Discover: generate enterprise ontologies from metadata; connect existing ontologies and semantic models; build AI-ready knowledge graphs.
- Ground: give agents trusted business meaning for metrics, entities, and relationships; reconcile inconsistent definitions across systems.
- Validate: check generated queries against certified definitions and lineage; return provenance with every answer.
- Expose: serve context to any agent harness through MCP, CLI, and API. Learn: fold certified corrections and usage back into the ontology, so shared context improves with every question.