Lilly and CDD Open AI Drug Discovery Toolbox for Biotech Innovators

BURLINGAME, Calif. – May 20, 2026 – A landmark agreement between Collaborative Drug Discovery (CDD) and pharmaceutical giant Eli Lilly and Company is set to democratize access to some of the most advanced artificial intelligence tools in drug discovery. CDD announced today it will integrate Lilly TuneLab, a sophisticated AI/ML platform, directly into its widely used CDD Vault data management solution, effectively placing big pharma's predictive power into the hands of smaller biotech companies.

This partnership will allow biotech firms using CDD Vault to access select predictive models trained on decades of Lilly's proprietary research data. For scientists working on the next generation of medicines, this means the ability to vet potential drug candidates faster, more accurately, and at a fraction of the cost previously required.

"By integrating TuneLab directly into CDD Vault, we are advancing CDD's core vision to enable collaboration across drug discovery teams and organizations," said Barry A. Bunin, CDD Chief Executive Officer and President. "We believe that solving the most complex challenges in drug discovery will depend on innovative collaboration models that provide broad access to research data and empower chemists and biologists to make informed, data-driven decisions."

Leveling the Playing Field in Drug Development

The collaboration addresses a long-standing disparity in the biopharmaceutical industry. For years, large pharmaceutical companies have invested billions to build massive internal datasets and develop proprietary AI models to streamline their R&D pipelines. This has created a significant competitive advantage, leaving smaller, more nimble biotech firms to rely on less sophisticated tools or more resource-intensive experimental methods.

The integration of Lilly TuneLab into CDD Vault aims to dismantle this barrier. The AI models, trained on what Lilly estimates to be over $1 billion in research investment, cover crucial aspects of early-stage development. The initial release will provide models for predicting a small molecule's ADMET properties (Absorption, Distribution, Metabolism, Excretion, and Toxicity) and assessing the "developability" of antibody therapeutics, such as their stability and solubility. Access to these predictions can help researchers identify flawed compounds early, reducing the risk of costly late-stage failures and potentially cutting early screening time by as much as 40-50%.

For a startup or mid-sized biotech, the ability to run these high-caliber predictive analyses within their existing workflow is a game-changer. It allows them to de-risk their projects, attract investment, and compete more effectively in a crowded market.

Lilly's Strategy: Collaboration Over Secrecy

Eli Lilly’s decision to share its prized AI models is a calculated strategic move, reflecting a broader industry trend toward open innovation. This initiative, part of Lilly's Catalyze360 program, is designed to foster external innovation that can ultimately accelerate scientific advancement across the board.

The core of this strategy lies in a sophisticated technology known as federated learning. Instead of requiring biotechs to upload their sensitive, proprietary compound data to an external server, the TuneLab integration allows Lilly's models to run securely within the biotech's own environment. Only anonymized model improvements are shared back, not the underlying chemical data. This clever architecture protects the intellectual property of all parties while creating a virtuous cycle: as more companies use the system, the models become more robust and accurate for everyone, including Lilly's internal teams.

This approach is a key part of Lilly's ambition to become an "AI-native" organization. The company has made massive investments in AI, including a multi-billion dollar deal with Insilico Medicine and an AI supercomputing lab with NVIDIA. By establishing TuneLab as a trusted industry tool—it was also recently integrated into competitor Benchling's platform—Lilly is positioning itself at the center of a new, collaborative AI ecosystem, subtly influencing discovery pipelines while accelerating its own R&D through shared learning.

Supercharging the Scientist's Workflow

For the individual chemist or biologist, the impact of this partnership will be felt daily. The integration is designed to be seamless, embedding powerful AI predictions directly into the CDD Vault platform where scientists already manage their experimental data.

Instead of being a separate, cumbersome tool, the Lilly models will complement CDD Vault’s existing suite of AI capabilities. These features are designed to work in concert, creating a powerful research engine. For instance, a scientist might use CDD's "Generative Bioisosteres" feature to create novel molecular ideas. They can then immediately run those new structures through Lilly's ADMET models to predict their behavior in the body and use CDD's integrated Boltz-2 module to dock them against a protein target—all within a single, secure software environment.

This synergy is a key focus for CDD. "TuneLab's models are synergistic with our innovations such as Zero Click Models, Generative Bioisosteres, as well as Ultrafast Deep Learning similarity to SureChEMBL for novelty and Enamine libraries for convenient SAR-by-catalog," shared CDD Research Informatics Senior Scientist Dr. Peter Gedeck.

This holistic approach transforms the research process from a linear, trial-and-error sequence into a dynamic, data-driven cycle of design, prediction, and validation. It empowers scientists to ask more complex questions and pursue more innovative therapeutic strategies with greater confidence.

A New Era of Collaborative Innovation

While the partnership between CDD and Lilly is a major headline, it also signifies a broader shift in the pharmaceutical landscape. The era of siloed R&D is giving way to a more interconnected ecosystem where collaboration and data sharing—facilitated by secure platforms—are seen as essential drivers of progress.

CDD, founded in 2004 with the vision of enabling web-based collaboration, is a natural hub for this new model. Its established reputation as a trusted, secure platform for managing sensitive chemical and biological data provides the foundation of trust necessary for such a partnership to succeed. By integrating Lilly's world-class AI, CDD is not just adding a feature but reinforcing its role as a central conduit for innovation.

The path forward involves navigating challenges related to cost, data standards, and regulatory acceptance of AI-driven data. However, this collaboration provides a clear and compelling blueprint for the future. By combining the agility and innovation of biotech with the scale and data resources of big pharma, such partnerships promise to accelerate the entire process of bringing life-saving medicines from the laboratory to the patients who need them most.