Instructions to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ", dtype="auto")

llama-cpp-python

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ",
	filename="MoQ-Quants/MoQ-3.0.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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ 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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
# Run inference directly in the terminal:
llama cli -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
# Run inference directly in the terminal:
llama cli -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
# Run inference directly in the terminal:
./llama-cli -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

Use Docker

docker model run hf.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

LM Studio
Jan
Ollama
How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Ollama:
```
ollama run hf.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
```

Unsloth Studio

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ 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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ 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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ to start chatting

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

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": "w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

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 w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

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 "w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16" \
  --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"

Docker Model Runner
How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Docker Model Runner:
```
docker model run hf.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16
```

Lemonade

How to use w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ:BF16

Run and chat with the model

lemonade run user.Gemma-4-12B-it-qat-GGUF-MoQ-BF16

List all available models

lemonade list

w-ahmad commited on Jun 10

Commit

535dbb7

verified ·

1 Parent(s): 08542f8

Update README.md

Browse files

Files changed (1) hide show

README.md +14 -0

README.md CHANGED Viewed

@@ -22,6 +22,20 @@ Results: At similar bits-per-weight (Bpw), MoQ outperforms Unsloth Dynamic quant
 "MoQ models are much better than UD quants on benchmarks, and they are also more token-efficient."
 <div align="center">
   <img src=https://ai.google.dev/gemma/images/gemma4_banner.png>
 </div>

 "MoQ models are much better than UD quants on benchmarks, and they are also more token-efficient."
+| Folder Link | BPW | Total Size | Description |
+| :--- | :---: | :---: | :--- |
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-3.0.gguf) | **3.0** | **4.06 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-3.25.gguf) | **3.25** | **4.81 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-3.5.gguf) | **3.5** | **5.05 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-3.75.gguf) | **3.75** | **5.63 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-4.0.gguf) | **4.0** | **6.38 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-4.25.gguf) | **4.25** | **6.46 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-4.5.gguf) | **4.5** | **6.54 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-4.75.gguf) | **4.75** | **6.61 GB**
+| [📂 **MoQ-Quants**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/MoQ-Quants/MoQ-5.0.gguf) | **5.0** | **7.57 GB**
+| [📂 **bf16**](https://huggingface.co/w-ahmad/Gemma-4-12B-it-qat-GGUF-MoQ/blob/main/bf16/Gemma-4-12B-it-qat-q4_0-unquantized-BF16.gguf) | **16.0** | **23.83 GB**
 <div align="center">
   <img src=https://ai.google.dev/gemma/images/gemma4_banner.png>
 </div>