Skip to content
Contact us

Federated Computing Provides Decentralized Intelligence for Drug Development At-Scale

Picture of Rhino

Today, we share exciting news on a groundbreaking application transforming how pharma can create living ecosystems of biotech partners, enabling AI model sharing, fine tuning, and inference – without sacrificing security, privacy, and governance – all while protecting valuable IP. 

Lilly TuneLab™ is Eli Lilly's new Federated AI program that lets biotechs securely run inference and fine-tune Lilly models on their own data - without moving data or IP. Now live with numerous biotech partners and 16 ready-to-use models available for immediate use across discovery and pre-clinical workflows, the program is quickly demonstrating how federated AI can shift the paradigm on what drug development partnership can and should look like.

Why It Matters

Drug R&D needs robust, real-world tested AI, not just demos. Setting the stage for fewer false starts, faster iteration, and more shots on goal, Lilly TuneLab is a giant leap forward in enabling faster selection of the right candidates - from bench to bedside - with greater confidence. 

How It Works

On the backend of Lilly TuneLab is the Rhino Federated Computing Platform (FCP), which deploys NVIDIA FLARE, and provides a flexible multi-cloud architecture with end-to-end data management and generative AI-powered workflows, federated dataset and computing apps, and federated trusted research environments (fTRE).  Also supporting the program are OpenAI’s ChatGPT for enabling the LLM-driven user experience and AWS with cloud and data analytics services. 

Together, Rhino FCP and the supportive technologies allow for:ith

  • Federated Inference

    Partners run Lilly proprietary models on their local proprietary data to generate decision-ready insights

  • Federated Fine-Tuning

    Partners contribute signals to improve predictive ADMET and other models without sharing raw data - to accelerate candidate evaluation and reduce reliance on animal testing.

  • Multi-Stakeholder Collaboration

    A scalable, collaborative ecosystem for creating multi-stakeholder partnerships, leveraging multi-modal data, for AI-driven drug discovery.


"AI only moves the needle when it meets real data, real constraints and real partners. TuneLab turns that belief into practice - helping biotechs design better candidates and helping Lilly select the right ones, faster. " -  Dr. Ittai Dayan, CEO, Rhino Federated Computing

Of note, as part of Lilly TuneLab, startups gain access to NVIDIA Inception, a global program designed to nurture companies revolutionizing industries with AI and accelerated computing. 

Become a Partner

This is a living ecosystem that grows more valuable with every partner and use case. We’ll be adding more models, workflows, and federated use cases in the coming months.

 

Stay Tuned

Today, we celebrate this momentous milestone, but as Lilly TuneLab grows, we’ll be sharing more on its progress and success.

Firefly_Gemini Flash_navy rhino icon 63593

 

Share your thoughts