
Continues KERMT molecular pretraining on a custom SMILES corpus inside a GPU container
- Open source
- 4K GitHub stars
- Health and science


Continues KERMT molecular pretraining on a custom SMILES corpus inside a GPU container

Pretrains a KERMT molecular model from scratch on your own SMILES corpus

Runs predictions with a finetuned KERMT checkpoint on a SMILES-only CSV

Finetunes a pretrained KERMT molecular encoder on your labeled SMILES CSV in Docker

Extracts per-molecule embeddings from KERMT checkpoints in a containerized GPU run

Fine-tunes public CodonFM Encodon checkpoints on labeled coding-sequence data

Generates and optimizes small molecules from a seed SMILES using NVIDIA's MolMIM NIM

Sets up the public CodonFM repo, dev container and Encodon checkpoint downloads

Validates coding-sequence CSVs and extracts frozen CodonFM embeddings for mRNA modeling

Runs masked-codon variant scoring with the public CodonFM Encodon models

Fine-tunes Parakeet TDT ASR on clinical vocabulary and measures the gain offline

Guides writing and debugging GPU-accelerated RDKit cheminformatics code with nvMolKit
The closest alternatives to Kermt add cmim pretrain are Kermt continue pretrain, Kermt pretrain scratch, Kermt infer, Kermt finetune and Kermt embed.
Yes. Kermt continue pretrain, Kermt pretrain scratch, Kermt infer have a free or open-source option.