Nemotron 3 Super

NVIDIA· released 11 Mar 2026

nvidia/nemotron-3-super-120b-a12b

Input / 1M tokens: $0.08
Output / 1M tokens: $0.45

Context window262K tokens

Longest answer: 236K tokens

reads textopen weightstool callingreasoning

About

Nemotron 3 Super is an open-weight NVIDIA text model with a hybrid Mixture-of-Experts design. It has 120B parameters and activates 12B parameters. The listed summary describes it as intended for complex multi-agent applications, and its architecture combines Mamba and Transformer elements. It accepts text and produces text, with tool and reasoning support. The context window is 262,144 tokens, and maximum output is 235,929 tokens. Its Hugging Face identifier is nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8. Input costs USD 0.08 per 1M tokens and output costs USD 0.45 per 1M tokens. The model was released on 2026-03-11; its ID is nvidia/nemotron-3-super-120b-a12b.

Who it is for

It suits teams working on text-based multi-agent applications that need tools or reasoning. Open weights and the listed context window may also matter when evaluating it for a stack.

What is good

Open weights
Supports tools and reasoning
262,144-token context window
Maximum output of 235,929 tokens

What to know first

Text input and output only
Input costs USD 0.08 per 1M tokens
Output costs USD 0.45 per 1M tokens

Verdict

Nemotron 3 Super combines open weights, tool use and reasoning support with a large context window. Its listed prices are USD 0.08 per 1M input tokens and USD 0.45 per 1M output tokens.

Details

Lab: NVIDIAopenrouter.ai · 3 Oct 2026
Context: 262Kopenrouter.ai · 3 Oct 2026
Input price: $0.08 / 1Mopenrouter.ai · 3 Oct 2026
Output price: $0.45 / 1Mopenrouter.ai · 3 Oct 2026
Max output: 235,929 tokensopenrouter.ai · 3 Oct 2026
Inputs: textopenrouter.ai · 3 Oct 2026
Open weights: Yesopenrouter.ai · 3 Oct 2026

More newest models

See the list

Ling 3.1 Flash ctx262KinFreeoutFree inclusionAI Apodex 1.1 Mini (free) ctx262KinFreeoutFree Apodex Pareto 26.10 Preview ctx1,049Kin$0.8 / 1Mout$3.2 / 1M unbiased GPT-6.1 Sol ctx1,050Kin$2 / 1Mout$10 / 1M OpenAI GPT-6.1 Sol Pro ctx1,050Kin$2 / 1Mout$10 / 1M OpenAI Claude Sonnet 5.5 ctx1,000Kin$2 / 1Mout$10 / 1M Anthropic

Listed on Inferse

Newest Models in 2026466 listed
Longest-Context Models in 2026466 listed
Cheapest Language Models in 2026437 listed
Models with Tool Calling in 2026395 listed
Reasoning Models in 2026332 listed
Open-Weight Models in 2026178 listed
NVIDIA Models: Pricing per 1M Tokens and Context (2026)11 listed

Sources

openrouter.ai/nvidia/nemotron-3-super-120b-a12b· checked 3 Oct 2026