AI & Machine Learning
Nemotron Customize
Nemotron Customize orchestrates Nemotron‑native steps—curation, translation, SFT/PEFT, pretraining, RL alignment, benchmark evaluation, checkpoint conversion, and ModelOpt optimization—into single‑step or multi‑step pipelines. It evaluates trained checkpoints or hosted endpoints, supports environment profiles, and allows data scientists and AI engineers to clean, translate, train, fine‑tune, align, convert, optimize, and compose workflows efficiently. The skill is tailored exclusively for Nemotron‑specific tasks and does not provide frontend, generic ML advice, billing, or non‑Nemotron coding.
- Publisher / source
- NVIDIA Skills (Verified)
- Upstream record
- NVIDIA/skills
- Version
- Version 2
- Published
- Publication review
- Approved for publication
- Content fingerprint
985a97bb839e8f031a7a7c1bba67c54372fdef80ded896ba45a0b43ce83c5672