Models / NVIDIA
Code
Chat

NVIDIA Nemotron 3 Super

Hybrid MoE model optimized for multi-agent workflows on single GPU deployment.

This model is not available on Together’s Serverless API.

  • Model provider
    NVIDIA
  • Type
    Code
    Chat
  • Main use cases
    Chat
    Reasoning
    Coding Agents
  • Parameters
    120B
  • Activated parameters
    12B
  • Context length
    1M
  • Input modalities
    Text
  • Output modalities
    Text
  • Category
    Chat