Nemotron 3 Super
A 120B-total, 12B-active reasoning model for coding, long-context, and agent workflows. NVIDIA publishes the weights, training datasets, and development recipes.
Models
- Nemotron 3 Super NVIDIA.
A 120B-total, 12B-active reasoning model for coding, long-context, and agent workflows. NVIDIA publishes the weights, training datasets, and development recipes.