Nemotron 3 Super

Name: Nemotron 3 Super
Author: NVIDIA

NemotronCurrent

NVIDIA(enterprise)

Released March 11, 2026

Run this model yourself API Docs Playground

Context262.1K tokens

CategoryLarge Language Model

About Nemotron 3 Super

Nemotron 3 Super is NVIDIA's 120B-parameter model with 12B active parameters. Its default Hugging Face configuration uses 256K context; NVIDIA documents up to 1M model context with additional VRAM. Hosted endpoint limits may differ.

Capabilities

textcode

Input Modalities

text

Output Modalities

text

Technical Details

API Identifier: nvidia/nemotron-3-super-120b-a12b:free
Category: Large Language Model
Context Window: 262,144 tokens

Run this model yourself

Local and cloud setup for Nemotron 3 Super, with exact configurations, sources and clear evidence limits.

Find a model for your hardware

No reviewed setup guide is available for this model yet. Explore configurations with documented support in the hardware finder.

Pricing and access

Free tier available on OpenRouter with rate limits

NVIDIAenterprise

The Engine of AI

2 Tools6 ModelsFounded 1993Santa Clara, California, United States

View full profile

2 Tools·6 Models

Tools by NVIDIA

Other AI tools from the same organization.

OpenShell

AI Infrastructure

OpenShell - Safe Runtime for Autonomous AI Agent Workflows

NeMo

Developer Tools

NeMo - NVIDIA Framework for Generative AI Builders

AI Models by NVIDIA

AI models from the same organization.

Model	Context Window	Price (In / Out per M)	Capabilities
Nemotron 3.5 Lightning 30B A3BCurrent	1.0M	$0.10 / $0.95	textreasoningcodingtool use+5
Nemotron 3 SuperCurrent	262K	$0.09 / $0.45	textcode
Nemotron 3 Nano 30B A3BCurrent	256K	-- / --	textcode
Nemotron 3 Nano 30B A3BCurrent	262K	$0.05 / $0.20	textcode
EagleCurrent	--	-- / --	visionimage understandingmultimodalresearch