Instructions to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-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 HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-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 HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Ollama
How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with Ollama:
```
ollama run hf.co/HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
```
Unsloth Desktop

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with Docker Model Runner:
```
docker model run hf.co/HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M
```

Lemonade

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-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 HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

NVIDIA Nemotron Labs 3 Elastic 23B A2.8B GGUF

A smaller, hacker-friendly GGUF build of NVIDIA Nemotron Elastic.

Built for llama.cpp, LM Studio, and other GGUF-compatible runtimes.

Files

File	Notes	Rough VRAM target
`nemotron-elastic-12b-Q4_K_S.gguf`	Smaller 4-bit quant	~10GB VRAM
`nemotron-elastic-12b-Q4_K_M.gguf`	Better 4-bit quant	~10GB+ VRAM

Which one should I use?

Use Q4_K_S if you want the easier/smaller 4-bit file.

Use Q4_K_M if you want the better-quality 4-bit file and have a little more room.

Both files are intended for roughly 10GB VRAM class hardware, depending on context size, KV cache settings, and GPU offload.

LM Studio

Open LM Studio and search for:

HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-23B-A2.8B-GGUF

Downloads last month: 213

GGUF

Model size

12B params

Architecture

nemotron_h_moe

Hardware compatibility

4-bit

Model tree for HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF

Base model

nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16

Finetuned

nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-NVFP4

Quantized

(3)

this model

Collection including HackerTwins/NVIDIA-Nemotron-Labs-3-Elastic-12B-A2B-GGUF

Nemotron-Labs-Elastic-12B

Collection

4-bit precision models for vLLM and llama.cpp • 1 item • Updated May 14