Google DeepMind pairs Antigravity agents with Cheminformatics Skills for drug discovery

On November 4, 2026, Google DeepMind announced the release of Cheminformatics Skills, a set of tools designed to work with the Google Antigravity agent to accelerate drug discovery workflows. The announcement came in an official post authored by Zoi Tsangalidou, Ruijie (Kelly) Zhang, Sukhdeep Singh and Michael V. LeVine.

The post describes designing new medicines as a process that involves curating complex data from charts, figures and literature, training predictive models, and proposing candidate molecules for synthesis. Google DeepMind says a single iteration of this workflow typically takes multiple days.

Using real-world lung cancer treatment data, the company says an agent running on Google Antigravity searched through thousands of compounds, performed matched molecular pair analysis, and proposed novel rigidified macrocycles complete with synthetic routes. According to Google DeepMind, the agent completed this in roughly one hour, compared with the multiple days the same iteration would conventionally require.

The announcement consists of Google DeepMind's own description of the system's capabilities, and the company has not yet released independent validation of the results or detailed benchmark methodology. The post does not state when the Cheminformatics Skills will be made available more broadly, nor does it compare the agent's output against other computational chemistry tools.