MENU

NVIDIA CUDA-X: AI-Defined Fab Tools and Industry Adopters

NVIDIA News USA
Overview
NVIDIA has expanded its CUDA-X library portfolio, including the new cuPotential for atomic-scale simulation, to enable “AI-defined fabs.” These tools accelerate chip design and manufacturing workflows for leaders like Synopsys and Cadence, reducing simulation times from months to a single day in some applications.
In Depth

Key Findings

NVIDIA is deploying a suite of CUDA-X libraries to automate semiconductor engineering and optimize “AI-defined fabs”—factories where AI optimizes processes and improves production efficiency. A primary addition is the cuPotential library, which allows developers to build AI agents that simulate atomic and molecular interactions, accelerating GPU-optimized workflows by an additional 2-3x in early-access testing.

Technical Details

The expanded CUDA-X portfolio provides specialized tools for different stages of the semiconductor lifecycle:

  • Simulation: cuPotential handles atomic-scale interactions; cuEST models electronic behavior; and cuDSS solves complex engineering simulations.
  • Data Processing: NanoVDB, an open-source 3D data library, has helped Lam Research achieve 30x faster simulation.
  • Inspection: The NVIDIA NPP image processing library is used by KLA to accelerate chipmaking inspection.
  • Lithography: The cuLitho library accelerates calculations for printing chip patterns.

Industry adoption shows significant performance gains: Synopsys reports potential 30x speedups using cuEST and 12x speedups using cuDSS. Cadence’s AuraStack AI Super Agent optimizes advanced packages for heat and stress 8x faster via cuDSS. Voltai’s Voltaire platform compresses complex electro-thermal-mechanical simulation workflows from months to a day.

Background & Context

As demand for advanced chips increases, manufacturing processes have become more complex. Engineers now require digital twins to model materials and test process changes before production. To support this, NVIDIA is integrating its Nemotron open models and NeMo with these libraries to enable “agentic AI”—AI agents capable of running and iterating on simulations autonomously.

Significance & Outlook

The shift toward AI-defined fabs aims to improve yield and production capacity. To measure progress toward fully autonomous semiconductor engineering, ChipAgents and NVIDIA are developing the Build-A-Chip Bench, a new AI benchmark that evaluates agents across end-to-end workflows from specification to graphic data system.

Source: https://nvidianews.nvidia.com/news/nvidia-and-semiconductor-leaders-advance-ai-across-chip-design-and-manufacturing-for-ai-defined-labs

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

  • Your email and selected fields are used only to deliver the newsletter.
  • We never share your information with third parties.
  • You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Takes about a minute · Unsubscribe anytime

Published by Troy-Technical, an independent site run by one engineer with a career in materials development.
About the author / Contact info@troy-technical.jp
Let's share this post !

Author of this article

TOC