NVIDIA officially unveiled the Nemotron 3 family, a new generation of open-source artificial intelligence models. These models are designed to power the development of sophisticated AI agents. The family features three sizes: Nano, Super, and Ultra.
Nemotron 3 is engineered to be significantly more efficient and powerful than previous versions. The most compact model, Nemotron 3 Nano, became available immediately. The more powerful Super and Ultra versions are planned for release in the first half of 2026.
This launch is a strategic move addressing increasing competition from open-source models developed by Chinese technology firms. NVIDIA also releases its training data and associated tools. This action fosters a transparent and secure platform for developers and enterprise clients.
The Nemotron 3 family has already seen early adoption. Major industry players adopting the models include Accenture, Oracle, Palantir, and ServiceNow, signaling strong initial interest.