All Models

Nemotron 3 Super

nemotron Reasoning Tool Calling Open Weights Structured Output

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...

Providers 3
Released Mar 11, 2026
Input Modalities text
Output Modalities text
Tarsk Use coding

Benchmarks

25.7
Intelligence Index
37.7
Coding Index

Available Providers (3)

Provider Model ID Input Cost Output Cost Context Max Output Docs
Nvidia nvidia/nemotron-3-super-120b-a12b $0.20/MTok $0.80/MTok 262.1K 262.1K
Weights & Biases nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 $0.20/MTok $0.80/MTok 262.1K 262.1K
Ollama Cloud nemotron-3-super —/MTok —/MTok 262.1K 65.5K

Capabilities

Reasoning
Tool Calling
Attachments
Open Weights
Structured Output