NVIDIA Nemotron 3.5 Super VL 120B A12B BF16 released
Source
huggingface.co
Author
NVIDIA model releases
Date
Why it matters
This is the reference-precision checkpoint of NVIDIA's open 120B multimodal mixture-of-expertsA model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.Full definition → (12B active), useful as the baseline for evaluation, 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 your own quantizationShrinking a model by storing its numbers less precisely — like rounding — so it runs faster and fits on smaller hardware, at a small quality cost.Full definition →.
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.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
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.
