Description
NVIDIA Nemotron 3 Super, a 120B open hybrid MoE model (12B active), supporting up to 1M tokens context window
Stats
18.2K Downloads
8 stars
Capabilities
Minimum system memory
Tags
Last updated
Updated 1 day agobyREADME
Nemotron 3 Super
General purpose reasoning and chat model trained by NVIDIA. Contains 120B total parameters with only 12B active at a time, using a hybrid LatentMoE architecture with Multi-Token Prediction layers for efficient high-throughput inference.
Features a reasoning toggle to enable or disable intermediate reasoning traces, with improved accuracy on complex queries when reasoning is enabled. Also features a low effort reasoning mode for faster, more compact reasoning. Includes native agentic capabilities for tool use, making it suitable for collaborative agents, IT ticket automation, RAG systems, chatbots, and other AI-powered applications. Supports multiple languages including English, Spanish, French, German, Japanese, Italian, and Chinese.
Supports a context length of 1M tokens.
Custom Fields
Special features defined by the model author
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Low Effort
: boolean
(default=false)
Controls whether the model uses low reasoning effort
Truncate Thinking History
: boolean
(default=true)
Controls whether thinking history will be truncated to save context space
Parameters
Custom configuration options included with this model
Sources
The underlying model files this model uses
Based on