Revvity and Lilly Unite to Democratize AI Drug Discovery for Biotechs
- $1 billion: Value of the experimental dataset used to train Lilly TuneLab's initial 18 AI models.
- 30%: Potential reduction in preclinical costs for biotech companies using predictive models.
- 2025: Year Lilly TuneLab was launched, marking a significant milestone in AI-driven drug discovery.
Experts view this collaboration as a transformative step toward democratizing AI-driven drug discovery, leveraging federated learning to enhance innovation while protecting intellectual property, ultimately benefiting both large pharmaceutical companies and smaller biotech firms.
Revvity and Lilly Unite to Democratize AI Drug Discovery for Biotechs
WALTHAM, MA – January 09, 2026 – In a significant move to broaden access to cutting-edge drug discovery tools, Revvity and Eli Lilly and Company have announced a major collaboration. The partnership will integrate Lilly’s high-quality predictive artificial intelligence models into Revvity’s Signals platform, creating a powerful new avenue for biotech companies to accelerate their research and development programs.
The collaboration centers on making Lilly TuneLab, a suite of advanced AI/ML models trained on decades of Lilly’s proprietary research data, available through Revvity’s recently launched Signals Xynthetica offering. By leveraging a federated learning framework, the initiative aims to build a scalable, secure ecosystem where collective intelligence can thrive, potentially leveling the playing field in the hyper-competitive race to develop new medicines.
A New Era of Collaborative Discovery
At the heart of this alliance is the fusion of Lilly's deep well of biological data with Revvity's robust informatics infrastructure. Lilly TuneLab, which launched in September 2025, represents a massive institutional investment, with its initial 18 models built upon an experimental dataset from hundreds of thousands of unique molecules, valued at over $1 billion. These models are designed to predict small-molecule properties and assess antibody developability, critical tasks in the early stages of drug discovery.
Through Revvity's "Models-as-a-Service" framework, these powerful predictive tools will now be delivered directly into the workflows of thousands of discovery teams who already use the Revvity Signals platform to manage their experimental data. This integration is designed to be seamless, allowing researchers to apply Lilly's sophisticated models to their own discovery programs without needing to build comparable AI systems from scratch—a task far beyond the reach of most startups and mid-sized firms.
“Federated learning represents one of the most powerful paths forward for AI in drug discovery, but it requires the right platform to succeed,” said Kevin Willoe, president of Revvity Signals Software, in the initial announcement. “By providing access to Lilly’s world-class predictive models through the Signals Xynthetica platform, we are creating a practical, secure way for organizations of all sizes to contribute to, and benefit from, collective intelligence.”
The Federated Future: Solving the Data Privacy Puzzle
The partnership’s strategic linchpin is its reliance on federated learning, a technology that addresses the single greatest barrier to collaboration in pharmaceutical R&D: data privacy and intellectual property protection. In a traditional AI model, vast amounts of data are pooled in a central location for training. In the fiercely competitive and IP-sensitive world of drug discovery, this is often a non-starter.
Federated learning flips that model on its head. Instead of moving data, the model moves. Participating organizations can use the Lilly TuneLab models on their own private, secure data. The models learn from this new data locally, and only the anonymized mathematical learnings—the model updates—are sent back to improve the central model. The underlying proprietary compound structures and raw experimental data never leave the owner's secure environment.
This approach not only protects sensitive IP but also fosters a virtuous cycle of improvement. As more organizations participate, the central models become more robust and accurate for everyone, trained on a more diverse set of chemical and biological data than any single company could amass. While the approach is not without technical challenges, such as managing data heterogeneity and ensuring model convergence across disparate datasets, it represents a crucial step toward building a truly collaborative R&D ecosystem.
Leveling the Playing Field for Biotech Innovators
The most immediate impact of this collaboration is expected to be felt by the small and mid-sized biotech companies that form the backbone of therapeutic innovation. These firms often possess novel biological insights and creative approaches but lack the resources to leverage AI at the scale of a pharmaceutical giant. Access to TuneLab could help them de-risk their programs, prioritize the most promising compounds, and avoid costly late-stage failures by making better predictions earlier. Industry analyses suggest that effective use of such predictive models could reduce preclinical costs by up to 30%.
Recognizing the financial barriers to AI adoption, Lilly and Revvity are jointly funding access for selected participants. This support package includes access to Revvity's Signals One and Signals Xynthetica software as well as modeling credits, effectively lowering the barrier to entry. To qualify, startups must typically have a drug candidate in preclinical development, ensuring that participants are positioned to both benefit from and contribute valuable training data to the federated network.
This initiative is a key component of Lilly's broader "Catalyze360" program, which aims to support the life sciences ecosystem. Lilly has already established TuneLab partnerships with companies like Insitro and Circle Pharma, as well as integration agreements with leading software providers Benchling and Schrodinger, signaling a clear strategy to embed its AI tools across the industry.
Strategic Stakes for Industry Giants
While the collaboration is framed around democratizing access, the strategic calculus for both Lilly and Revvity is clear and compelling. For Lilly, the initiative is a long-term investment in its own R&D engine. By facilitating the improvement of its TuneLab models through federated learning, Lilly ensures its internal discovery teams have access to increasingly powerful and accurate predictive tools. It also provides the company with a bird's-eye view of emerging innovation, potentially identifying future partnership or acquisition opportunities.
For Revvity, the partnership solidifies its position as a critical infrastructure provider for the age of AI-driven drug discovery. By becoming the conduit to Lilly's highly sought-after models, Revvity significantly enhances the value proposition of its Signals platform, which includes tools for wet-lab data capture (Signals One) and secure external collaboration (Signals Synergy). This move is likely to drive adoption and expand its customer base, particularly within the high-growth biotech sector.
With the market for AI in drug discovery projected to reach tens of billions of dollars in the coming decades, this alliance is more than a simple technology partnership. It is a strategic positioning that could define new standards for how drugs are discovered, creating an ecosystem where shared intelligence, rather than siloed data, becomes the primary driver of therapeutic breakthroughs. The success of this model could ultimately influence the entire industry's approach to R&D, pushing it toward a more collaborative and efficient future.
📝 This article is still being updated
Are you a relevant expert who could contribute your opinion or insights to this article? We'd love to hear from you. We will give you full credit for your contribution.
Contribute Your Expertise →