Nvidia bets open models can give enterprise AI agents the context they’re missing
Nvidia announced its Agent Knowledge Foundation, a new addition to its Agent Toolkit designed to address a key challenge for enterprises: enabling AI agents to understand and work with company data across multiple systems. The system bundles Nemotron Retriever models, Auto-Ontology library, Kumo models for structured data, and CUDA-X processing libraries. The approach keeps data in place while allowing agents to gain enterprise context, addressing the gap between simple vendor connections and true cross-system understanding.
Key points
- Nvidia's Agent Knowledge Foundation bundles open-source AI models and data processing components
- Designed to give enterprise AI agents broader context without competing with existing data platforms
- Data remains under original access controls and management rules
Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. The New Stack did the reporting; read the original for the full story.
Read the full story at The New Stack
