NVIDIA’s Nemotron-4 15B Dominates Multilingual Domain, Defeating 4× Larger Rivals

In a new paper Nemotron-4 15B Technical Report , an NVIDIA research team introduces Nemotron-4 15B. Nemotron-4 15B is comprising 15 billion parameters, is trained on an extensive corpus of 8 trillion text tokens, showcasing unparalleled multilingual capabilities among models of comparable size.

ai artificial intelligence billion capabilities deep-neural-networks domain large language model machine learning machine learning & data science ml multilingual nvidia nvidia research paper parameters report research research team team technical technology text tokens