Content Paint

Nemotron 3: NVIDIA goes wide open on big-context, mixture-of-experts model

The chipmaker opened its Transformer-Mamba MoE model, plus a trove of training data, in hopes it will "help strengthen an open ecosystem" for AI.

Phillip de Wet

Dec 15, 2025 - 2 min read

Image credit: https://unsplash.com/@jogara_0

Get the full story: Subscribe for free

Join peers managing over $100 billion in annual IT spend and subscribe to unlock full access to The Stack’s analysis and events.

Already a member? Sign in

Success! You now have access to additional content.