# NVIDIA Nemotron 3 Super 120B A12B BF16

> NVIDIA Nemotron 3 Super 120B A12B BF16 is a large language model trained by NVIDIA for agentic workflows, long-context reasoning, and tool use. It employs a hybrid Latent Mixture-of-Experts (LatentMoE) architecture with 120B total parameters and 12B active parameters, configurable reasoning mode, and support for English, French, German, Italian, Japanese, Spanish, and Chinese.

Provider: NVIDIA · Type: LLM · State: Ready

model path: `accounts/fireworks/models/nemotron-3-super-120b-a12b-bf16`

[Deploy on demand](https://app.fireworks.ai/dashboard/deployments/create?baseModel=accounts/fireworks/models/nemotron-3-super-120b-a12b-bf16) · [Fine-tune](https://app.fireworks.ai/dashboard/fine-tuning/create?baseModel=accounts/fireworks/models/nemotron-3-super-120b-a12b-bf16)

NVIDIA Nemotron 3 Super 120B A12B BF16 is a large language model trained by NVIDIA for agentic workflows, long-context reasoning, and tool use. It employs a hybrid Latent Mixture-of-Experts (LatentMoE) architecture with 120B total parameters and 12B active parameters, configurable reasoning mode, and support for English, French, German, Italian, Japanese, Spanish, and Chinese.

## NVIDIA Nemotron 3 Super 120B A12B BF16 API Features

- **Fine-tuning**: NVIDIA Nemotron 3 Super 120B A12B BF16 can be customized with your data to improve responses. Fireworks uses LoRA to efficiently train and deploy your personalized model · [Docs](https://docs.fireworks.ai/fine-tuning/fine-tuning-models)
- **On-demand Deployment**: On-demand deployments allow you to use NVIDIA Nemotron 3 Super 120B A12B BF16 on dedicated GPUs with Fireworks' high-performance serving stack with high reliability and no rate limits. · [Docs](https://docs.fireworks.ai/guides/ondemand-deployments)

## Metadata

- **State**: Ready
- **Created on**: 20 MAY, 2026
- **Kind**: Base model
- **Provider**: NVIDIA
- **Huggingface**: [nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16)

## Specifications

- **Calibrated**: No
- **Mixture-of-Experts**: No
- **Parameters**: 120B

## Supporting Functionality

- **Fine-tuning**: Supported
- **Serverless**: Not Supported
- **Context Length**: 262k tokens
- **Function Calling**: Supported
- **Embeddings**: Not Supported
- **Rerankers**: Not Supported
- **Supporting image input**: Not Supported
