Roche is deploying more than 3,500 NVIDIA Blackwell GPUs across hybrid cloud and on‑premises environments in the U.S. and Europe, expanding on an existing collaboration to turn AI and accelerated computing into a core operating capability across its pharmaceutical and diagnostics businesses.
At NVIDIA GTC today, Roche unveiled how this new AI computing backbone will power everything from biological foundation models to drug discovery and digital twins for manufacturing.
This investment positions Roche with the extensive computing capacity it needs to enable scientists across the enterprise to pursue their most ambitious scientific projects.
Delivering an Enterprise-Grade AI Computing Backbone for Healthcare
“We’re excited to innovate at the intersection of science and technology to accelerate drug and diagnostic solutions development,” said Wafaa Mamilli, chief digital and technology officer at Roche. “With high-quality data and smarter AI, we will be able to leverage those insights both in pharma as well as in our diagnostic divisions.”
Roche’s combined on-premises and cloud GPU infrastructure now exceeds 3,500 high-performance GPUs — the greatest announced GPU footprint available to a pharmaceutical company. This hybrid architecture lets teams train massive models while supporting local development across global sites.
Roche is embedding accelerated computing into the foundation of how the company discovers, develops, manufactures and delivers healthcare solutions, moving AI from isolated pilots to an enterprise‑wide capability.
Harnessing Drug Discovery for Breakthroughs at a New Scale
AI is already transforming Roche and Genentech’s discovery engine, and the new infrastructure will accelerate this process further.
“When we talk about collapsing time, we’re really talking about the patients and their families who are waiting,” said Mamilli. “As an innovation-led company, using AI for this work is deeply meaningful and purpose-driven; it is exactly what we are here to do.”
In drug discovery, AI is at the heart of Genentech’s Lab-in-the-Loop strategy, where experiments, data and AI are connected in an iterative loop to tackle the hardest problems. Nearly 90% of Genentech’s eligible small‑molecule programs integrate AI. In one program, a degrader molecule for oncology was designed 25% faster, and in another program, AI helped deliver a backup molecule — a secondary drug candidate developed in parallel to mitigate costly risks of failure of the first one — in seven months instead of more than two years.
With NVIDIA Blackwell and the NVIDIA BioNeMo platform running on its AI factories, Roche can train and fine‑tune biological and molecular foundation models, integrate proprietary datasets and expand AI‑driven lab automation, all powering the Lab-in-the-Loop. This allows Roche to explore far larger swaths of the biological and chemical space — and to do it with greater speed.
Driving Pharmaceutical Manufacturing Powered by Digital Twins
Roche is also modernizing pharmaceutical manufacturing with NVIDIA Omniverse libraries, building digital twins of production facilities to simulate and optimize complex systems before they go live. These virtual environments are already helping accelerate the development of Roche’s new GLP‑1 manufacturing facility in North Carolina.
“We are using NVIDIA Omniverse to build high-fidelity digital twins of our manufacturing sites such as the facility in North Carolina,” said Mamilli.
AI is also being developed for regulatory documentation, quality assurance and production scheduling — areas where even small efficiency gains can ripple across a global supply chain.
Providing Diagnostics and Digital Health at Global Scale
Roche’s diagnostics business gives it a second engine for AI innovation, including in Roche’s digital health portfolio, where AI can help make lab operations more efficient, provide better clinical decision support solutions and help patients make better informed choices for their healthcare.
Accelerated computing and NVIDIA Parabricks software enable insights across vast datasets. In digital pathology, NVIDIA technologies scan a large number of images to detect subtle disease patterns. In digital health, Roche uses NVIDIA NeMo Guardrails to ensure safe and reliable healthcare-grade AI.
Roche is also evaluating the BioNeMo platform for developing AI models to support molecular design and chemistry innovation related to sequencing technologies.
Digital pathology is another frontier where Roche is advancing AI models to analyze pathology images and improve diagnostic accuracy at scale.
Expanding Access to Accelerated Computing Across Global Organization
Roche is undergoing one of the most significant AI infrastructure deployments in the pharmaceutical and diagnostics industry.
By expanding access to accelerated computing across its global organization, the company is shifting AI from a specialized discipline to a core capability embedded in everyday scientific and clinical workflows.
The impact spans the full healthcare continuum: more viable drug targets, faster therapeutic development, deeper integration of diagnostics and therapeutics, more efficient manufacturing and new forms of precision medicine that can reach patients worldwide.
Roche’s infrastructure scale-up is the foundation for its next decade of innovation — from larger biological foundation models and agentic research workflows to AI‑driven lab automation and advanced digital diagnostics.
“Everything we do here is going to allow us to get that medicine and those diagnostic devices and solutions faster to the patients,” said Mamilli.
Learn more about NVIDIA AI in healthcare and life sciences.
Categories:
AI Infrastructure
Related News
