GenMol

Discrete-diffusion generative molecule design and optimization

GenMol generates and optimizes molecules with a discrete state-space diffusion framework that supports fragment editing, inpainting and multi-parameter optimization. It handles scaffolding, R-group decoration and linker design within a single model. Targets property constraints such as potency proxies, QED and synthetic accessibility.

GenMol is being onboarded — request access and be first in line.