Instructions to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf",
	filename="gemma-4-12b-qat-it-assistant-Q4_0-Q6emb.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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
# Run inference directly in the terminal:
llama cli -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
# Run inference directly in the terminal:
llama cli -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
# Run inference directly in the terminal:
./llama-cli -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

Use Docker

docker model run hf.co/RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

LM Studio
Jan
Ollama
How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with Ollama:
```
ollama run hf.co/RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
```

Unsloth Studio

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf to start chatting

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

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": "RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

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 "RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0" \
  --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 RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with Docker Model Runner:
```
docker model run hf.co/RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0
```

Lemonade

How to use RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RachidAR/gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf:Q4_0

Run and chat with the model

lemonade run user.gemma-4-12B-it-qat-q4_0-MTP-assistant-gguf-Q4_0

List all available models

lemonade list

Gemma 4 12B QAT MTP Assistant GGUF

GGUF quantizations converted from google/gemma-4-12B-it-qat-q4_0-unquantized-assistant.

Tested with llama.cpp b9549+ (Gemma 4 MTP support).

Recommendations

Q4_0_Q4emb — recommended for most users
Q8_0 — highest precision

Files

gemma-4-12b-qat-it-assistant-Q4_0-Q6emb.gguf
gemma-4-12b-qat-it-assistant-Q4_0_Q4emb.gguf
gemma-4-12b-qat-it-assistant-Q8_0.gguf

Notes

Q4_0_Q4emb keeps token_embd.weight in Q4_0 rather than the higher-precision embedding quantization normally used by llama.cpp.

Initial testing showed similar draft acceptance rates while reducing file size.

Example

llama-server \
  -m gemma-4-12B-it-qat-Q4_0.gguf \
  -md gemma-4-12b-qat-it-assistant-Q4_0_Q4emb.gguf \
  --spec-type draft-mtp \
  --spec-draft-n-max 2