Box Integrates Nvidia Nemotron Embedding System for Enhanced Content Retrieval
At its recent developer conference, Box announced 'Box Content Retrieval,' a new feature that integrates Nvidia’s Nemotron embedding models into its core product. This integration is designed to improve the efficiency of AI agents by utilizing embedded metadata to process queries more effectively. According to Nvidia, the system is expected to deliver query responses four times faster while reducing errors by 20% and token usage by 90%. The technology leverages Nemotron's ability to map underlying concepts rather than relying solely on keywords, which significantly improves search relevance for complex tasks like legal research. The feature will be rolled out to all Enterprise Advance customers early next year, combining Nvidia’s open-weight models with Box’s proprietary metadata and permission controls.
Topics: Cloud Content Management Enterprise Software Generative AI Nvidia Nemotron