Models/NVIDIA/Nemotron 3 Super

Nemotron 3 Super

Name: Nemotron 3 Super API access
Brand: NVIDIA

NVIDIA · Open weights · Released Mar 11, 2026 · nvidia/nemotron-3-super-120b-a12b

Compare Estimate cost

1M contextToolsReasoningStructured outputPrompt cachingOpen weights

Available from 2 providersUSD / MTOK · VERIFIED 4 MINUTES AGO

EST. MONTHLY COSTclick a row for detail · click headers to sort

PROVIDER	INPUT	OUTPUT	CACHE	CONTEXT	QUANT	TTFT	TPS	UPTIME
DeepInfra	$0.085	$0.40	—	262K	bf16	2.2s	—	100.0%
DIDigitalOcean	$0.30	$0.65	$0.06	1M	—	1s	15 tok/s	99.8%

TTFT/TPS: median (P50) over trailing 72h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · methodology v1.1 · measured 44 hours ago

Pricing variants · USD / MTok in / out

STANDARD

$0.085 / $0.40

per MTok in / out

FREE

Free / Free

per MTok in / out

About

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...

Specifications

Context window1,000,000

Max output16,384

Modalities intext

Modalities outtext

LicenseOpen weights

ReleasedMar 11, 2026

TokenizerOther

WeightsHugging Face ↗

Open weights

Hugging Facenvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 ↗

Downloads154.5K

Likes275

Licenseother

safetensorspytorch

Nemotron 3.5 Lightning$0.08 / $0.20 Nemotron 3 Ultra$0.50 / $2.20 Nemotron 3 Nano 30B A3B$0.05 / $0.20

Embed badge

[![Nemotron 3 Super](https://modelindex.ai/badge/nvidia/nemotron-3-super-120b-a12b)](https://modelindex.ai/models/nvidia/nemotron-3-super-120b-a12b)

Sources

OpenRouter API4 minutes ago

AGGREGATOR

Hugging Face4 minutes ago

COMMUNITY

ModelIndex benchmarks44 hours ago

MEASURED

Verified 4 minutes ago · list prices, not negotiated rates.