30B / 3B active (MoE)
NVIDIA Open Model License
Open 30B MoE model with 3B activated parameters designed for strong reasoning and agentic capabilities in a text-generation interface. Official release notes describe it as post-trained from NVIDIA-Nemotron-3-Nano-30B-A3B-Base and operating in both thinking and instruct (non-reasoning) modes. The release highlights strong competition-level results on math and code reasoning tasks (including IMO/IOI-oriented evaluations and LiveCodeBench-family metrics), alongside broad coverage across STEM, alignment, and long-context evaluations. The model card links openly released supervision artifacts for reproducibility, including SFT and RL datasets, and provides benchmark comparisons against related open models in the same parameter class.