Box taps Nvidia’s Nemotron for new agent-first content retrieval system

Box adopts Nvidia Nemotron to power agentic content retrieval platform

By GPU Data Hub Desk · GPU Data Hub·Editorial standards·5 Nov 2026, 14:00

Enterprise software vendor Box has integrated Nvidia's Nemotron model into a newly developed information retrieval system tailored for autonomous agents. The updated architecture is claimed to lower token consumption by 90 per cent whilst simultaneously accelerating response times and lowering error rates.

Original source

This summary was written by the GPU Data Hub desk from reporting published by TechCrunch on 5 Nov 2026, 14:00. Read the full article at the original publisher.

Read Original Source →

Companies

Box Nvidia

Topics

enterprise AI content retrieval LLMs agentic AI

LinkedIn X

More from GPU Data Hub

Latest →

The hidden supply chain costs of the AI boom

Harvard Business School academic Willy Shih has highlighted the complex manufacturing and economic pressures underpinning worldwide logistics during the artificial intelligence expansion. The analysis examines the less obvious infrastructural and production expenditure necessary to sustain rapid technological adoption.

AI Infrastructure·2h ago·The Verge

Congresswoman calls for artificial superintelligence ban after Democrats' seismic midterm sweep

United States Representative Yassamin Ansari of Arizona has urged her party to push for a ban on artificial superintelligence following recent midterm elections. The congresswoman is advocating for lawmakers to direct their legislative focus toward the severe risks posed by advanced frontier models.

AI Infrastructure·4h ago·Tom's Hardware

Anthropic loosens Claude's game cloning ban — AI will now offer to build original games in the requested genre

Anthropic has updated Claude's rules around game cloning, allowing the assistant to propose original software within a requested category rather than outright rejecting prompts. The revision aims to cut down on unwarranted blocks so that software creators can use the model for legitimate development workflows.

AI Infrastructure·6h ago·Tom's Hardware

Fitting AI Models To Edge Applications

Adapting artificial intelligence models for embedded devices requires establishing precise operational requirements before deployment. System architects typically deploy a combination of pruning, distillation, and quantization to shrink workloads without sacrificing performance. This coordinated compression approach is proving essential for running compact models and small language models directly on edge hardware.

AI Infrastructure·9h ago·Semiconductor Engineering

Why Testing Automated Driving Is Harder Than It Looks

Validating autonomous vehicle platforms poses significant challenges that real-world road trials struggle to address comprehensively. Engineers are increasingly deploying synthetic simulation environments to safely replicate uncommon road conditions and critical edge cases. This virtual testing model allows developers to thoroughly stress automated driving systems prior to physical deployment.

AI Infrastructure·9h ago·Semiconductor Engineering

NCSA Industry Conference: Driving Innovation with AI, Quantum, and Supercomputing

The National Center for Supercomputing Applications convened commercial partners, hardware specialists and academics at its recent conference marking four decades of operation. Attendees examined practical methods for deploying advanced computing, artificial intelligence and quantum resources whilst managing escalating operational expenditures.

AI Infrastructure·16h ago·HPCwire