NVIDIA

NVIDIA Nemotron 3 Super 120B A12B

Generally availableOpen weights262K contextReasoning

Summary

NVIDIA Nemotron 3 Super 120B A12B is an open-weight language model from NVIDIA, released on 11 Mar 2026. It accepts text and generates text, with a 262K-token context window and up to 131K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.07 per million input tokens and $0.40 per million output tokens (OpenRouter), across 4 providers we track.

Nemotron middle tier for collaborative agents and high-volume reasoning workloads

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperNVIDIA
Release date11 Mar 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsOpen weights [Hugging Face]
Licensenvidia-nemotron-open-model-license [Hugging Face]
Parameters124B [Hugging Face]
Context window262K (262,144 tokens) [models.dev]
Max output131K tokens [models.dev]
Knowledge cutoffJun 2025 [models.dev]
InputText [models.dev]
OutputText [models.dev]
Reasoning modeYes [models.dev]
Tool callingYes [models.dev]
Structured outputYes [models.dev]
Hugging Facenvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 [Hugging Face]

NVIDIA Nemotron 3 Super 120B A12B API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
OpenRouter
nvidia/nemotron-3-super-120b-a12b
$0.07$0.40–262KOpenRouter, 18 Nov 2026
Amazon Bedrock
nvidia.nemotron-super-3-120b
$0.15$0.65–262Kmodels.dev, 18 Nov 2026
Vercel AI Gateway
nvidia/nemotron-3-super-120b-a12b
$0.15$0.65–256Kmodels.dev, 18 Nov 2026
Nebius AI Studio
nvidia/nemotron-3-super-120b-a12b
$0.30$0.90$0.06262Kmodels.dev, 18 Nov 2026

Running NVIDIA Nemotron 3 Super 120B A12B locally

Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.

QuantizationWeightsTotal, 8K contextTotal, 32K context
Q4_K_M (4-bit)75 GB76 GB76 GB
Q8_0 (8-bit)131 GB132 GB132 GB
FP16 / BF16 (16-bit)247 GB248 GB248 GB

At 4-bit with an 8K conversation, NVIDIA Nemotron 3 Super 120B A12B needs about 76 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.

Frequently asked questions

What is the context window of NVIDIA Nemotron 3 Super 120B A12B?

NVIDIA Nemotron 3 Super 120B A12B has a context window of 262,144 tokens (262K) and can generate up to 131,072 tokens in a single response, according to models.dev.

How much does the NVIDIA Nemotron 3 Super 120B A12B API cost?

NVIDIA Nemotron 3 Super 120B A12B is listed from $0.07 per million input tokens and $0.40 per million output tokens (OpenRouter). Prices last verified 18 Nov 2026.

Is NVIDIA Nemotron 3 Super 120B A12B open source?

NVIDIA Nemotron 3 Super 120B A12B is an open-weight model: its weights are published on Hugging Face as nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 under the nvidia-nemotron-open-model-license license. Check the license terms before commercial use.

When was NVIDIA Nemotron 3 Super 120B A12B released?

NVIDIA released NVIDIA Nemotron 3 Super 120B A12B on 11 Mar 2026, according to models.dev.

Which providers offer NVIDIA Nemotron 3 Super 120B A12B?

We track NVIDIA Nemotron 3 Super 120B A12B on 4 providers: OpenRouter, Amazon Bedrock, Vercel AI Gateway, Nebius AI Studio.

How much VRAM does NVIDIA Nemotron 3 Super 120B A12B need?

About 76 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 132 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Gemma 4 31B ITGoogle DeepMind2 Apr 2026262K$0.14$0.40Open
Gemma 4 26B A4B ITGoogle DeepMind2 Apr 2026262K$0.07$0.30Open
Step 3.5 FlashStepFun29 Jan 2026262K$0.10$0.30Open
Mistral Small 4Mistral AI16 Mar 2026262K$0.15$0.60Open
Jamba MiniAI21 Labs8 Jan 2026256K$0.20$0.40Open
GLM-4.7-FlashZhipu AI (Z.ai)19 Jan 2026200K$0.06$0.40Open

Change history

  • Context window changed from 256K to 262K
  • Max output changed from 32K to 131K
  • Tool calling changed from no to yes
  • Max output changed from 16K to 32K
  • Context window changed from 262K to 256K
  • Tool calling changed from yes to no
  • Name changed from Nemotron 3 Super 120B A12B to NVIDIA Nemotron 3 Super 120B A12B
  • Max output changed from 131K to 16K
  • Name changed from NVIDIA Nemotron 3 Super 120B A12B to Nemotron 3 Super 120B A12B
  • Max output changed from 33K to 131K
  • Name changed from Nemotron-3-Super-120B-A12B to NVIDIA Nemotron 3 Super 120B A12B
  • Name changed from NVIDIA Nemotron 3 Super 120B A12B to Nemotron-3-Super-120B-A12B
  • Max output changed from 131K to 33K
  • Max output changed from 33K to 131K
  • Name changed from Nemotron-3-Super-120B-A12B to NVIDIA Nemotron 3 Super 120B A12B
  • Name changed from Nemotron 3 Super 120B A12B to Nemotron-3-Super-120B-A12B
  • Max output changed from 16K to 33K
  • Tool calling changed from no to yes
  • Context window changed from 256K to 262K
  • Max output changed from 32K to 16K
  • Name changed from NVIDIA Nemotron 3 Super 120B A12B to Nemotron 3 Super 120B A12B
  • Context window changed from 262K to 256K
  • Max output changed from 33K to 32K
  • Tool calling changed from yes to no
  • Name changed from Nemotron-3-Super-120B-A12B to NVIDIA Nemotron 3 Super 120B A12B

Sources