@nemotron-nano-3
NVIDIA's efficient large language model optimized for long-context reasoning, code, and multi-document analysis.