NVIDIA: Nemotron 3 Super
NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...
Everything else
| Max output | 16K |
|---|---|
| Input / 1M | $0.085 |
| Output / 1M | $0.4 |
| Cache read / 1M | — |
| Cache write / 1M | — |
| 1M in + 1M out | $0.485 |
| Takes | text |
| Returns | text |
| Tools | yes |
| Reasoning | optional · medium, low |
| Open weights | yes |
| Released | 2026-03-11 |
| Knowledge cutoff | — |
| Retires | — |
| Catalog id | nvidia/nemotron-3-super-120b-a12b |
Using it
In 00, this model is picked per agent — and per tier, so the model that answers your customers need not be the one that writes your code. Get 00.