Kermt add cmim pretrain vs Kermt continue pretrain
Kermt add cmim pretrain is an open-source agent skill for health and science. Kermt continue pretrain is an open-source agent skill for health and science.
- Kermt add cmim pretrain is best for extend an existing grover encoder checkpoint with a contrastive objective.
- Kermt continue pretrain is best for continue pretraining a kermt model on my own smiles library.
- What it is
- Kermt add cmim pretrain is an open-source agent skill for health and science.Kermt continue pretrain is an open-source agent skill for health and science.
- Best for
- Extend an existing GROVER encoder checkpoint with a contrastive objectiveContinue pretraining a KERMT model on my own SMILES library
- Kind
- SkillSkill
- Pricing
- Open sourceOpen source
- Platforms
- Claude Code, CodexClaude Code, Codex
- GitHub stars
- 4K4K
- Last commit
- Nov 2026Nov 2026
- License
- Apache-2.0Apache-2.0
- Open source
- Good for
- Extend an existing GROVER encoder checkpoint with a contrastive objective; Continue pretraining a molecular model on a custom SMILES dataset; Convert a legacy grover_base checkpoint for hybrid training without retraining from scratchContinue 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 training
What Kermt add cmim pretrain does
Checkpoint upgrade to hybrid
Adds a randomly initialized cMIM decoder and latent head to a GROVER encoder checkpoint.
Checkpoint validation
Rejects inputs that already have contrast heads or task FFN heads before any work begins.
Corpus preparation
Validates the pretraining CSV and builds a fresh SMILES vocabulary sized to it.
Continued hybrid pretraining
Launches a detached training run in a container that trains vocabulary and contrastive objectives together.
Reproducible run records
Stores the upgrade summary and run metadata so the upgraded checkpoint's provenance can be audited.
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.
Kermt add cmim pretrain vs Kermt continue pretrain: questions
Is Kermt add cmim pretrain or Kermt continue pretrain cheaper?
Both are open source. Compare current plans on each official site.
What is the difference between Kermt add cmim pretrain and Kermt continue pretrain?
Kermt add cmim pretrain is an open-source agent skill for health and science. Kermt continue pretrain is an open-source agent skill for health and science.
