---
title: "Nvidia's Nemotron Embedding System Integrated into Box's Core Product"
url: https://noti.group/nvidia-s-nemotron-embedding-system-integrated-into-box-s-core-product/
language: en
publisher: "Noti Group"
section: "Technology"
published: 2026-11-05T14:00:00.000Z
updated: 2026-11-05T17:38:28.078Z
id: 40c5f405-ab14-4a0c-a06a-afb35368435a
source: "TechCrunch https://techcrunch.com/2026/11/05/box-taps-nvidias-nemotron-for-new-agent-first-content-retrieval-system/"
attribution: "Link to https://noti.group/nvidia-s-nemotron-embedding-system-integrated-into-box-s-core-product/ and name Noti Group when you quote or summarize this story."
---

# Nvidia's Nemotron Embedding System Integrated into Box's Core Product

Box has unveiled a new feature that integrates Nvidia's Nemotron embedding system into its core product, called Box Content Retrieval.

Box, a leading cloud content management company, has unveiled a new feature that integrates Nvidia's Nemotron embedding system into its core product. The new feature, called Box Content Retrieval, is designed to enhance the performance of agents in answering queries by leveraging embedded metadata created by Nemotron.

The benefits for users are clear: faster query resolution times, improved accuracy, and reduced costs. According to Kari Briski, Nvidia's VP of Generative AI software, this new feature will enable queries to be answered four times faster with 20% fewer errors and a significant reduction in tokens used.

Box Content Retrieval is set to become available to all Enterprise Advance customers early next year. This means that users can expect to see enhanced search functionality from the company's agents, who will be able to provide more relevant results without having to sift through each document individually.

The integration of Nemotron into Box's product is an implementation of Nvidia's Embed project, which aims to improve automated retrieval systems using open-weight embedding models. These models rely on Large Language Models (LLMs) to pre-process data and make it easier for agents to work with.

Nvidia's recent release of Nemotron in July demonstrated the potential of this approach, ranking #1 on the RTEB benchmark, a measure of retrieval quality that has often been overlooked by major labs.

The new system developed by Box is expected to be most effective when combined with more advanced agents that can interact seamlessly with the Nemotron-processed data.

This means that professionals such as lawyers will benefit from the system's ability to quickly locate relevant clauses and precedents, even if they are described using different language. The underlying concepts embedded in the Nemotron models enable related legal ideas to be surfaced, making it easier for users to find what they need.

The system relies on both the freely available Nemotron models and Box's existing metadata and permission controls. This combination of technologies can make it difficult to recreate the system from scratch, with five different components integrated into a single API.

As a result, the new content retrieval system is poised to offer significant advantages in terms of efficiency and accuracy, particularly for users who require access to complex or specialized information.

The new content retrieval system developed by Box in collaboration with Nvidia's Nemotron technology is poised to offer significant advantages in terms of efficiency and accuracy. This is particularly relevant for users who require access to complex or specialized information.

The system leverages the power of Nemotron to enable advanced search capabilities, allowing users to retrieve specific content more quickly and accurately than traditional methods. With this new system, Box aims to enhance user experience and improve productivity by streamlining the process of finding and accessing relevant information. The impact is expected to be substantial, with improved efficiency being a major benefit for users who rely on complex or specialized knowledge.

---
Source: [TechCrunch](https://techcrunch.com/2026/11/05/box-taps-nvidias-nemotron-for-new-agent-first-content-retrieval-system/)  
Published by Noti Group: https://noti.group/nvidia-s-nemotron-embedding-system-integrated-into-box-s-core-product/