TLDR: Eli Lilly and Nvidia have announced a groundbreaking partnership to build the pharmaceutical sector’s most powerful AI supercomputer, an ‘AI factory,’ designed to dramatically accelerate drug discovery and development. This initiative, featuring over 1,000 Nvidia Blackwell Ultra GPUs and projected to be operational by January, aims to reduce the time and cost of bringing new medicines to patients by enabling millions of virtual experiments and leveraging large-scale AI models.
Pharmaceutical giant Eli Lilly and Company has officially partnered with tech leader Nvidia Corporation to construct what is being hailed as the pharmaceutical industry’s most powerful AI supercomputer. This ambitious collaboration aims to establish an ‘AI factory’ that will revolutionize drug discovery and development, significantly shortening timelines and reducing associated costs. The supercomputer, a first-of-its-kind Nvidia DGX SuperPOD configured with DGX B300 systems, is set to incorporate over 1,000 Nvidia Blackwell Ultra GPUs. It is anticipated to deliver more than 9,000 petaflops of AI performance, enabling a vast array of workloads from genome-scale analyses to the training of foundational models using Lilly’s extensive experimental data. The system is expected to be fully operational by January 2026.
This ‘AI factory’ will manage the entire AI lifecycle, encompassing data acquisition, training, fine-tuning, and high-volume inference, thereby facilitating AI-based research at an unprecedented scale. The core objective is to accelerate the identification of new molecules and streamline the typically lengthy drug development process. Beyond initial discovery, Eli Lilly plans to deploy this immense computing power in various critical areas. These include optimizing clinical trials through AI-assisted document workflows, advancing precision medicine, improving production efficacy by creating ‘digital twins’ of manufacturing lines for predictive maintenance, and discovering new biomarkers through medical-imaging research.
Diogo Rau, Lilly’s chief information and digital officer, emphasized the transformative potential, stating, “I don’t believe any other company in our industry is doing what we do at this scale.” He added, “As a 150-year-old medicine company, one of our most powerful assets is decades of data. With purpose-built models and AI, we can set a new scientific standard that accelerates innovation to deliver medicines to more patients, faster.” Thomas Fuchs, chief AI officer at Lilly, further elaborated on the strategic shift: “Lilly is shifting from using AI as a tool to embracing it as a scientific collaborator. By embedding intelligence into every layer of our workflows, we’re opening the door to a new kind of enterprise: one that learns, adapts, and improves with every data point.”
Also Read:
- NVIDIA Unveils AI Factory Blueprint to Bolster U.S. Government Operations and National Security
- Nvidia Unveils Sweeping Full-Stack AI Strategy: Powering Supercomputers and Pioneering 6G Networks
Eli Lilly also intends to make select proprietary AI drug discovery models available to partners via Lilly TuneLab, a machine learning (ML) platform developed at an approximate cost of $1 billion. This federated platform, utilizing Nvidia’s FLARE framework, will allow biotech companies to fine-tune models without needing to share their underlying data, provided they contribute ‘training data’ to refine these models. This new supercomputer is poised to surpass existing systems in the pharmaceutical sector, including Recursion’s BioHive-2, which previously held the title of the largest supercomputer owned by a pharma company. Notably, the new facility will operate on 100% renewable electricity within existing company facilities and will employ chilled-water liquid cooling, underscoring a commitment to sustainability. This initiative is part of Lilly’s broader $50 billion commitment to expanding its U.S. manufacturing and R&D footprint, reinforcing its position as a leader in AI innovation within the pharmaceutical industry, as recognized by CB Insights.
