DigestAI news desk

Cut through the AI noise.

Agents & Tools4 min read

NVIDIA releases Auto Ontology, an open semantic layer for AI agents

NVIDIA announced Auto Ontology, an open‑source library under the Apache 2.0 license that provides a semantic layer between AI agents and data warehouses. The framework builds an ontology from enterprise metadata and lets agents generate SQL against certified business definitions. The default LLM is nvidia/nemotron-3-nano-30b-a3b, with Nemotron embedding and reranking models, and it can run via…

1 source

Key points

  • NVIDIA’s Auto Ontology library cuts ontology creation time by 90% in its own testing.
  • Default model is nvidia/nemotron-3-nano-30b-a3b; the stack runs on Docker Compose or Kubernetes via Helm.
  • The MCP server provides nine read‑only tools, but no public text‑to‑SQL accuracy scores have been published.

NVIDIA’s own testing claims the tool cuts ontology‑creation time by 90%, though no public text‑to‑SQL accuracy scores have been released. The MCP server exposes nine read‑only tools, including askquestion that returns answer, SQL and rows, with a 100‑row cap. An evaluation harness ships with datasets such as BIRD Dev (1,534 questions, 11 databases) and Mini‑Dev (500 questions), but the benchmark relies on deterministic execution rather than LLM judges. Users can integrate any OpenAI‑compatible endpoint, and the source code, containers and docs are available on GitHub.

Full story from MarkTechPost · by Asif RazzaqOpen source ↗

NVIDIA Releases Auto Ontology: An Open Semantic Layer That Grounds AI Agent SQL

MarkTechPost · 2 November 2026

Loading the full article…

This text was published by MarkTechPost and written by Asif Razzaq. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
NVIDIAnvidia/nemotron-3-nano-30b-a3b

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Agents & Tools

All →

Related stories