Chemistry & ADMET · Tsinghua University
Self-supervised molecular GNN, strong across ADMET endpoints.
No backend is configured for this model.
1–30s
Not runnable yet
This model isn't available to run right now.Run this model
About this model
GROVER pre-trains a graph transformer on 10 million unlabelled molecules with node-, edge- and graph-level objectives. Graph structure beats string representation on many ADMET tasks, and GROVER is where that shows most clearly.
Standardized I/O contract
Every model in the Hub speaks the same contract, which is what lets the Router and the agent call any of them without special-casing.
Inputs
Ligand SMILES
Outputs
ADMET endpoint predictions.
