Rockwell Automation: Advancing Industrial Intelligence with Edge-Based Generative AI Using NVIDIA Nemotron

Author photo: Craig Resnick
ByCraig Resnick
Category:
Company and Product News

This release outlines how Rockwell Automation’s integration of Nvidia Nemotron Nano is designed to address emerging industrial intelligence requirements across secure, real-time environments.

Rockwell Automation announced that it is bringing its generative AI capabilities directly to the industrial edge through its integration of the open-source NVIDIA Nemotron-Nano-9B-v2 model. The small language model is designed to be optimized for FactoryTalk Design Studio and additional Rockwell Automation product workflows, helping to support real-time intelligence and more efficient decision-making across industrial teams.

To address the need for more reliable intelligence in secure environments, Rockwell Automation is applying Nemotron Nano distillation techniques to create an SLM that operates with lower compute, space and power requirements. By fine-tuning the model with data used for FactoryTalk Design Studio Copilot, it is building a capability designed for industrial automation personnel.

Key elements of this approach include:

  • Use of the open-source NVIDIA Nemotron-Nano-9B-v2 model and the NVIDIA NeMo framework.

  • Distillation techniques helping to enable more efficient edge operation.

  • Fine-tuning with existing FactoryTalk Design Studio Copilot datasets.

  • A focus on predictable performance in more constrained industrial environments.

Built for use across design, development, production and maintenance workflows, the model operates on a range of environments. Its deployment flexibility is designed to help support industrial teams working across distributed or air-gapped operations.

Supported environments include:

  • HMI panels.

  • Industrial appliances.

  • Desktop integrated development environments.

  • Server and private cloud systems.

  • Edge and offline deployments, including air-gapped networks.

Early evaluations have reported performance improvements in reasoning, responsiveness and parallel processing, helping to better align with industry momentum toward lighter weight AI models that offer more predictable, real-time support at the point of work. These developments help to reflect broader sector trends toward edge-ready intelligence, localized compute and strict data control across industrial operations.

Rockwell Automation plans to demonstrate these capabilities as part of its focus on industrial intelligence and edge-based AI workloads, better aligning with the growing shift toward real-time contextual decision support across operations.

Follow ARC Advisory Group for the latest trends in sustainability, automation, and next generation technology.

Engage with ARC Advisory Group

Representative End User Clients
Representative Automation Clients
Representative Software Clients