by Nvidia

Nemotron 3 Super.

text open weights datacenter 120B params 1M ctx
Cheapest input
$0.075/M
on DeepInfra
Cheapest output
$0.4/M
on DeepInfra
Smallest GPU
2× Nvidia Q RTX 8000
$0.54/hr
Description

About Nemotron 3 Super.

An open-weights text model from NVIDIA with 120B parameters, 12B of them active per token. It uses a LatentMoE design that combines Mamba-2, mixture-of-experts and attention layers, plus a built-in multi-token prediction head for faster decoding. Context runs up to 1M tokens, and reasoning can be switched on or off. NVIDIA aims it at agent workflows, long-context reasoning, tool use, retrieval and high-volume jobs such as IT ticket automation, and lists 8x H100 80GB as the minimum GPU setup.

Context window

How much it can remember.

1M tokens ≈ 750,000 English words
4K 32K 128K 1M
Capabilities

What it can do.

· Vision input
· Audio input
· Video input
· Function calling
· Tool use
· JSON mode
✓ Streaming
· Fine-tuning