NVIDIA Nemotron 3.5 Super VL 120B A12B Base BF16 released
Source
huggingface.co
Author
NVIDIA model releases
Date
Why it matters
The base checkpoint is the starting point for your own fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition → or post-training of the Nemotron 3.5 Super family, with the pretrainingThe first, biggest phase of building a model: training it on enormous amounts of text so it learns language, facts, and reasoning in general.Full definition → dataset referenced in the card.
Terms in this piece · Glossary
mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
quantization — Shrinking a model by storing its numbers less precisely — like rounding — so it runs faster and fits on smaller hardware, at a small quality cost.
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
pretraining — The first, biggest phase of building a model: training it on enormous amounts of text so it learns language, facts, and reasoning in general.
