Variational AI, a generative AI drug discovery company, has announced a collaboration with Merck to apply its Enki™ platform to design and optimize novel small molecule candidates against two therapeutic targets.

Under the terms of the agreement, Variational AI will use a fine-tuned version of its platform, trained on Merck’s proprietary data, to generate and optimize these small molecules. In return, Variational AI will receive an upfront payment and is eligible for milestones with a total potential value of up to $349 million. Merck will have the exclusive right to develop and commercialize any compounds that arise from the partnership. According to Handol Kim, the CEO of Variational AI, this framework has the potential to significantly accelerate and redefine the economics of drug discovery. Dr. Robert M. Garbaccio, Vice President and Head of Discovery Chemistry at Merck Research Laboratories, expressed excitement about working with Variational AI to harness the potential of AI to improve efficiency and speed in the early stages of the drug discovery process.

Variational AI's Enki™ platform is built on a foundation model that uses a curated version of all publicly available data, as well as proprietary generative models, to create and optimize small molecule leads based on partner target profiles.

The company, founded by machine learning researchers from institutions like MIT and Google Research, aims to design novel, optimized small molecules with an improved probability of success to accelerate the development of transformative therapeutics.