All Models

nvidia nemotron 3 super 120b a12b

nemotronReasoningTool CallingStructured Output

NVIDIA Nemotron 3 Super is a hybrid Mixture-of-Experts (MoE) model engineered for highest compute efficiency and accuracy in multi-agent applications and specialized agentic systems. It is optimized to run many collaborating agents per application on a single GPU, delivering high accuracy for reasoning, tool use, and instruction following.

Providers1
ReleasedMar 11, 2026
Input Modalitiestext
Output Modalitiestext
Tarsk Usecoding

Benchmarks

13.6
Intelligence Index
37.7
Coding Index

Available Providers (1)

ProviderModel IDInput CostOutput CostContextMax OutputDocs
Requestynvidia-nemotron-3-super-120b-a12b$0.10/MTok$0.50/MTok262.1K262.1K

Capabilities

Reasoning
Tool Calling
Attachments
Open Weights
Structured Output