Bristol Myers Squibb builds the most powerful AI cluster in life sciences based on NVIDIA Vera Rubin

Bristol Myers Squibb (BMS) has announced the deployment of its second NVIDIA DGX SuperPOD, built on eight DGX Vera Rubin NVL72 systems, each featuring NVIDIA Vera CPUs and Rubin GPUs. The new cluster delivers up to 10x performance per megawatt compared to the previous infrastructure and provides access to a unified AI platform, including the BioNeMo Agent Toolkit for biological AI. BMS Vice President of Research Erin Davis emphasized that now every scientist at the company can use the supercomputer without restrictions. BMS has been using the first DGX SuperPOD for three years, achieving significant results: AI-driven target identification saves weeks of manual work, and the CELMoD compound library for targeted degradation of cancer proteins has been expanded thanks to AI. Additionally, the "Predict First" methodology in lead optimization allows filtering out molecules that do not meet target properties, increasing the efficiency of laboratory experiments. The new cluster will be integrated into a unified environment accessible from any BMS site, managed via NVIDIA Mission Control. Agentic workflows will enable data analysis across programs and projects, which, according to Senior Vice President Payal Sheth, creates a cumulative learning cycle that did not exist before. Davis noted that tasks are already scheduled on the new cluster, ranging from small and large molecule design to clinical applications and digital twins.