NVIDIA Nemotron 3 Super 120B A12B (Q3_K_XL) — 15.4 GBon NVIDIA RTX 4080

Q3_K_XL NVIDIA RTX 4080

Overview

NVIDIA Nemotron 3 Super 120B A12B is a 123.61B parameter moe language model by NVIDIA, with code, multilingual, thinking, tool-calls capabilities. It supports a context window of up to 262,144 tokens.

Nemotron 3 Super 120B A12B is a 123.61-billion-parameter hybrid Mamba-2 Transformer LatentMoE model from NVIDIA, activating 12 billion parameters per token across 22 of 512 routed experts plus 1 shared expert. Trained on over 25 trillion tokens, it targets agentic reasoning, code generation, tool calling, and multilingual conversation in 7 languages. A 256K context window, toggleable thinking mode, and multi-token prediction enable high-throughput inference for complex multi-agent workflows. Its MoE sparsity quantizes well to GGUF for self-hosted deployment on multi-GPU setups.

At Q3_K_XL quantization (low quality tier), the model weighs 58.33 GB. This exceeds the 16 GB of VRAM on NVIDIA RTX 4080. Inference is still possible via CPU offload or memory-mapped loading from disk, but expect significantly reduced performance.

The NVIDIA GeForce RTX 4080 is a consumer GPU with 16 GB of GDDR6X VRAM and 717 GB/s memory bandwidth. It delivers 48.7 FP16 TFLOPS, providing capable performance for local LLM inference on mid-size models. It handles quantized models up to 13B parameters effectively. A good fit for developers who want Ada Lovelace performance at a more accessible price point.

Hardware Requirements

Model size	58.33 GB
VRAM available	16 GB
VRAM used	15.4 GB
Min RAM required	58.2 GB
GPU layers	88 / 88
Context size	167,206
Backend	cuda13
Flash attention	Yes

Performance Notes

Mixture-of-Experts

This is a Mixture-of-Experts model with 512 total experts, of which 22 are active per token. Only the active experts need fast GPU access — inactive expert weights can reside in system RAM with minimal impact on generation speed.

Read the MoE deployment guide →

Possible performance impact

Expert weights offloaded to CPU via tensor override. At least 58.2 GB system RAM recommended — if insufficient, data will be read from disk on demand.

Shorter conversations

Context size has been reduced due to memory constraints. This may affect the model's ability to handle long conversations.

Deploy

Server Local

Prerequisites

Ensure your GPU nodes are prepared with the NVIDIA container toolkit:

ansible-playbook prositronic.infra.nvidia_container_toolkit

Command

helmfile --state-values-file <(curl -s https://www.prositronic.eu/values/nemotron-3-super-120b-a12b/q3_k_xl/nvidia-rtx4080.yaml) apply

Generated values.yaml

/values/nemotron-3-super-120b-a12b/q3_k_xl/nvidia-rtx4080.yaml

Loading values…

Install llama.cpp

Install llama.cpp from the official build instructions for your platform and backend.

Download Model

curl -L -o nemotron-3-super-120b-a12b.gguf "https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF/resolve/main/UD-Q3_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q3_K_XL-00001-of-00003.gguf"

Start Server

llama-server \
  -m nemotron-3-super-120b-a12b.gguf \
  --n-gpu-layers 88 \
  --ctx-size 167206 \
  --flash-attn \
  -ot ".ffn_.*_exps.=CPU"

Verify

curl http://localhost:8080/health

Explore

• Configure a deployment
• Browse models
• Browse hardware

Other quantizations

• NVIDIA Nemotron 3 Super 120B A12B (Q8_0)
• NVIDIA Nemotron 3 Super 120B A12B (Q5_K_M)
• NVIDIA Nemotron 3 Super 120B A12B (Q4_K_M)

Other hardware

• NVIDIA Nemotron 3 Super 120B A12B on CPU Only
• NVIDIA Nemotron 3 Super 120B A12B on NVIDIA A100 80GB
• NVIDIA Nemotron 3 Super 120B A12B on NVIDIA H100

Frequently Asked Questions

How much VRAM does NVIDIA Nemotron 3 Super 120B A12B (Q3_K_XL) need?

The Q3_K_XL quantization of NVIDIA Nemotron 3 Super 120B A12B requires 58.33 GB. All 88 layers fit in the 16 GB of VRAM available on NVIDIA RTX 4080, enabling full GPU acceleration.

Can I run NVIDIA Nemotron 3 Super 120B A12B on NVIDIA RTX 4080?

Yes. NVIDIA RTX 4080 provides 16 GB of VRAM, which is enough to run NVIDIA Nemotron 3 Super 120B A12B (Q3_K_XL) with all layers on the GPU for optimal performance.

What is quantization?

Quantization reduces a model's numerical precision from its original floating-point format to a more compact representation. This shrinks the file size and VRAM footprint, making it possible to run large models on consumer hardware. The trade-off is a small reduction in output quality. Q3_K_XL compresses NVIDIA Nemotron 3 Super 120B A12B from its original size down to 58.33 GB.

What quantization should I choose for NVIDIA Nemotron 3 Super 120B A12B?

Q3_K_XL is a low-quality quantization. Higher-quality quants (Q8, Q6) preserve more model accuracy but need more VRAM. Lower quants (Q4, Q3, Q2) reduce VRAM usage at the cost of some quality. Choose based on your available hardware and quality requirements.

What is flash attention and why is it enabled?

Flash attention is a memory-efficient algorithm that speeds up the attention mechanism in transformer models. It reduces VRAM usage during inference by avoiding the materialisation of the full attention matrix. For NVIDIA Nemotron 3 Super 120B A12B on NVIDIA RTX 4080, flash attention is enabled to maximise context length and throughput within the available 16 GB of VRAM.

What is MoE and how does it affect deployment?

NVIDIA Nemotron 3 Super 120B A12B uses a Mixture-of-Experts (MoE) architecture with 512 experts, of which 22 are active per token. This means only a fraction of the model weights are used for each inference step, allowing MoE models to be larger in total parameter count while remaining efficient at inference time.

Last updated: March 12, 2026