Instructions to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF",
	filename="IQ3_S/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-IQ3_S-00001-of-00003.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Ollama:
```
ollama run hf.co/AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
```

Unsloth Studio

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF to start chatting

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Docker Model Runner:
```
docker model run hf.co/AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M
```

Lemonade

How to use AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AesSedai/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.NVIDIA-Nemotron-3-Super-120B-A12B-GGUF-Q4_K_M

List all available models

lemonade list

Updates

03/12/2026

I uploaded the wrong splits for Q4_K_M / Q5_K_M and have corrected that now with the changes mentioned in the 03/11 update. Also added an IQ3_S quant now that there is a PR from @bartowski to fix the IQ4_NL quantization crash.

03/11/2026

I've adjusted the Q4_K_M and Q5_K_M to use Q5_0 for the ffn_down_exps tensor, which brings the Q5_K_M quant size down substantially.

Description

This repo contains specialized MoE-quants for NVIDIA-Nemotron-3-Super-120B-A12B-BF16. The idea being that given the huge size of the FFN tensors compared to the rest of the tensors in the model, it should be possible to achieve a better quality while keeping the overall size of the entire model smaller compared to a similar naive quantization. To that end, the quantization type default is kept in high quality and the FFN UP + FFN GATE tensors are quanted down along with the FFN DOWN tensors.

Notes

This model is a little weird, architecturally. There isn't a ffn_gate_exps tensor in it, and the ffn_down_exps tensor has 2688 elements in it which means that it is not compatible with most Q*_K quantizations.

So you may notice that the ffn_down_exps here is a little odd, and producing an actual IQ3_S-sized quant like I normally do is tricky since the IQ4_NL quantization type is also not behaving well.

I've chosen to upload these 3 quants for now and hope that there will be some improvements soon.

Quant	Size	Mixture	PPL	1-(Mean PPL(Q)/PPL(base))	KLD
Q5_K_M	80.27 GiB (5.71 BPW)	Q8_0 / Q5_K / X / Q5_0	4.590127 ± 0.027865	+0.0817%	0.007533 ± 0.000042
Q4_K_M	73.70 GiB (5.25 BPW)	Q8_0 / Q4_K / X / Q5_0	4.600659 ± 0.027947	+0.3113%	0.010532 ± 0.000072
IQ4_XS	63.45 GiB (4.52 BPW)	Q8_0 / IQ3_S / X / Q4_1	4.647848 ± 0.028308	+1.3402%	0.022996 ± 0.000191
IQ3_S	52.66 GiB (3.75 BPW)	Q6_K / IQ2_S / X / IQ4_NL	4.787999 ± 0.029268	+4.3960%	0.059260 ± 0.000528