Nemotron 3 Super 120b A12b

NVIDIA logo

Nemotron 3 Super 120b A12b

Nemotron 3 Super 120b A12b

Nemotron 3 Super by NVIDIA is a hybrid Mamba-Transformer MoE model with 120B total and 12B active parameters, optimized for agentic reasoning, multi-step task planning, and long-context workloads.

MODALITIES

CONTEXT

1000000

INPUT PRICE

$0.00

ARTIFICIAL ANALYSIS CODING

37.7

ARTIFICIAL ANALYSIS INTELLIGENCE

25.4

PROVIDERS / REGIONS

All live provider offers for this model, with region and token pricing.

Provider

Product

Context

Location

Input

Cached input

Artificial Analysis Coding

Artificial Analysis Intelligence

No provider offers available.

Additional provider routing data is not available yet.

CAPABILITIES

Supported inputs, outputs, and model features from the live CMS record.

Text input

Tool calling

Reasoning

Streaming

Sampling parameters

Text modality

Text to speech input

Completion input

Completion output

JSON mode

Functions

Explore the full model garden

Compare providers, capabilities, and pricing across the live Orq.ai catalog.

Build on sovereign EU infrastructure