Kermt continue pretrain vs Kermt finetune
Kermt continue pretrain is an open-source agent skill for health and science. Kermt finetune is an open-source agent skill for health and science.
- Kermt continue pretrain is best for continue pretraining a kermt model on my own smiles library.
- Kermt finetune is best for finetune a pretrained molecular encoder on my solubility csv.
- What it is
- Kermt continue pretrain is an open-source agent skill for health and science.Kermt finetune is an open-source agent skill for health and science.
- Best for
- Continue pretraining a KERMT model on my own SMILES libraryFinetune a pretrained molecular encoder on my solubility CSV
- Kind
- SkillSkill
- Pricing
- Open sourceOpen source
- Platforms
- Claude Code, CodexClaude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, Goose
- GitHub stars
- 4K4K
- Last commit
- Nov 2026Nov 2026
- License
- Apache-2.0Apache-2.0
- Open source
- Good for
- Continue pretraining a KERMT model on my own SMILES library; Resume an interrupted molecular pretraining run from its last checkpoint; Check whether a checkpoint and CSV are valid before trainingFinetune a pretrained molecular encoder on my solubility CSV; Train a classification model on my labeled SMILES dataset with a scaffold split; Run multi-GPU finetuning on several ADMET endpoints
What Kermt continue pretrain does
Checkpoint validation
Checks that the input checkpoint is a supported pretrain type and redirects users when it is a finetuned or encoder-only file.
Vocabulary pass-through
Reuses the checkpoint's atom, bond and SMILES vocabulary files so the new corpus matches the existing model heads.
Containerized training
Launches the pretraining runner detached inside a container with mounted host directories for data, checkpoints and outputs.
GPU batch sizing
Chooses default batch sizes for single or multiple GPUs and offers a VRAM-based table for smaller hardware.
Optional released model fetch
Downloads the pinned NV-KERMT-70M-v2 bundle over HTTPS only with explicit consent or an opt-in flag.
Resumable runs
Supports fresh-schedule continuation or a true resume of an interrupted run with its saved schedule.
What Kermt finetune does
Checkpoint validation
Confirms the input is a pretrain checkpoint and not an already finetuned model before training begins.
Data and target checks
Validates the CSV structure and auto-detects numeric target columns for confirmation.
Feature preparation
Cleans SMILES, builds features and writes a manifest for the training step.
Detached containerized training
Launches single-GPU or multi-GPU DDP finetuning in a background container with logs and TensorBoard output.
Optional released model download
Fetches a pinned Hugging Face model bundle only after explicit user consent or a flag.
Replayable run manifest
Records the command, arguments and defaults used so a run can be reproduced.
Kermt continue pretrain vs Kermt finetune: questions
Is Kermt continue pretrain or Kermt finetune cheaper?
Both are open source. Compare current plans on each official site.
What is the difference between Kermt continue pretrain and Kermt finetune?
Kermt continue pretrain is an open-source agent skill for health and science. Kermt finetune is an open-source agent skill for health and science.
