NVIDIA-Nemotron-3-Super-120B-A12B

120B / 12B

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.

Context
262K
Input price
$0.09/1M tokens
Output price
$0.4/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming

Details

Model IDnvidia/NVIDIA-Nemotron-3-Super-120B-A12B
Provider DeepInfra
Creator NVIDIA
Familynemotron
License—
Statusactive
Typechat
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters120B / 12B
Architecture—
Context window262K tokens
Max input—
Max output—
Reasoning tokensYes
Knowledge cutoff—
Release date—
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.deepinfra.com/v1/openai/v1/chat/completions

Pricing

Input
$0.09
Output
$0.4
Cache write
—
Cache read
—
Batch in
—
Batch out
—

Family Comparison: nemotron

ModelContextMax outPricing
NVIDIA-Nemotron-Nano-9B-v2131K—$0.04$0.16
Nemotron-3-Nano-30B-A3B262K—$0.05$0.2
NVIDIA-Nemotron-3-Super-120B-A12B262K—$0.09$0.4
NVIDIA-Nemotron-Nano-12B-v2-VL131K—$0.2$0.6

API

GET/v1/models/deepinfra/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B