Nemotron 3
Nemotron-3 is a family of open-source large language models developed and released by NVIDIA. The family includes multiple model variants designed to serve different computational and deployment contexts, from resource-constrained environments to more capable configurations. All models in the family are made available under open-source licensing terms.
Nemotron-3 Nano
The Nemotron-3 Nano is a 30-billion-parameter variant within the Nemotron-3 family. This model size represents a balance between computational efficiency and task performance, making it suitable for deployment scenarios where moderate inference resources are available. Like other models in the family, it is released as an open-source resource for research and practical applications.