Instructions to use Pulsate1680/gemma-4-mtp-q8_0-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Pulsate1680/gemma-4-mtp-q8_0-gguf",
	filename="mtp-gemma-4-12B-it.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Pulsate1680/gemma-4-mtp-q8_0-gguf
# Run inference directly in the terminal:
llama-cli -hf Pulsate1680/gemma-4-mtp-q8_0-gguf

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Pulsate1680/gemma-4-mtp-q8_0-gguf
# Run inference directly in the terminal:
llama-cli -hf Pulsate1680/gemma-4-mtp-q8_0-gguf

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 Pulsate1680/gemma-4-mtp-q8_0-gguf
# Run inference directly in the terminal:
./llama-cli -hf Pulsate1680/gemma-4-mtp-q8_0-gguf

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 Pulsate1680/gemma-4-mtp-q8_0-gguf
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Pulsate1680/gemma-4-mtp-q8_0-gguf

Use Docker

docker model run hf.co/Pulsate1680/gemma-4-mtp-q8_0-gguf

LM Studio
Jan

vLLM

How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Pulsate1680/gemma-4-mtp-q8_0-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": "Pulsate1680/gemma-4-mtp-q8_0-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Pulsate1680/gemma-4-mtp-q8_0-gguf

Ollama
How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with Ollama:
```
ollama run hf.co/Pulsate1680/gemma-4-mtp-q8_0-gguf
```

Unsloth Studio

How to use Pulsate1680/gemma-4-mtp-q8_0-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 Pulsate1680/gemma-4-mtp-q8_0-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 Pulsate1680/gemma-4-mtp-q8_0-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Pulsate1680/gemma-4-mtp-q8_0-gguf to start chatting

How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Pulsate1680/gemma-4-mtp-q8_0-gguf

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": "Pulsate1680/gemma-4-mtp-q8_0-gguf"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Pulsate1680/gemma-4-mtp-q8_0-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 Pulsate1680/gemma-4-mtp-q8_0-gguf

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 Pulsate1680/gemma-4-mtp-q8_0-gguf

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with Docker Model Runner:
```
docker model run hf.co/Pulsate1680/gemma-4-mtp-q8_0-gguf
```

Lemonade

How to use Pulsate1680/gemma-4-mtp-q8_0-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Pulsate1680/gemma-4-mtp-q8_0-gguf

Run and chat with the model

lemonade run user.gemma-4-mtp-q8_0-gguf-{{QUANT_TAG}}

List all available models

lemonade list

Gemma 4 MTP Draft Models — GGUF Q8_0

This repository contains GGUF-converted MTP / assistant draft models for Gemma 4 speculative decoding in llama.cpp.

These files are not standalone chat models. They are draft / MTP models intended to be loaded alongside the matching Gemma 4 target model using llama.cpp speculative decoding.

Files

File	Source checkpoint	Source revision	GGUF type	SHA256
`mtp-gemma-4-31B-it.gguf`	`google/gemma-4-31B-it-qat-q4_0-unquantized-assistant`	`5db7ebef2cfabfc4b5b0dc898171d854b1521b14`	Q8_0	`c1a08236603ff83678494d491a80b941b5ed5dfed035ce778d662dfc435d832d`
`mtp-gemma-4-12B-it.gguf`	`google/gemma-4-12B-it-qat-q4_0-unquantized-assistant`	`9b957049a807d0ce8d7682d0f308a2df835c3f9a`	Q8_0	`13331068b6af643c3dc75e619373b674c1f75a1958e7c82e2020d96a17c63809`

What these are

Google's source checkpoints are unquantized QAT assistant checkpoints. They contain half-precision weights extracted from the QAT pipeline for Gemma 4 MTP / assistant drafting.

This repo provides those assistant checkpoints converted to GGUF and quantized as Q8_0 for use with llama.cpp's draft-mtp speculative decoding path.

The target model itself is not included in this repository. Pair these files with the matching Gemma 4 IT GGUF target model.

Conversion approach

Converted with llama.cpp's Hugging Face to GGUF converter:

python convert_hf_to_gguf.py \
  --outtype q8_0 \
  --outfile mtp-gemma-4-31B-it.gguf \
  /path/to/google/gemma-4-31B-it-qat-q4_0-unquantized-assistant

python convert_hf_to_gguf.py \
  --outtype q8_0 \
  --outfile mtp-gemma-4-12B-it.gguf \
  /path/to/google/gemma-4-12B-it-qat-q4_0-unquantized-assistant

Conversion environment:

llama.cpp commit: 7d2b45b4f
GGUF v3
Architecture: gemma4-assistant
Output type: Q8_0
49 tensors per model

A tokenizer config compatibility fix was applied before conversion because the local Transformers version errored on:

"extra_special_tokens": []

The value was changed to an empty object for conversion compatibility:

"extra_special_tokens": {}

No model weights were otherwise modified beyond GGUF conversion and Q8_0 quantization.

Usage with llama.cpp

Example for Gemma 4 31B:

llama-server \
  -m /path/to/gemma-4-31B_q4_0-it.gguf \
  -md /path/to/mtp-gemma-4-31B-it.gguf \
  --spec-type draft-mtp \
  --spec-draft-n-max 4 \
  --kv-unified \
  -fa on

For very large context sizes, the target model and KV cache may consume nearly all GPU memory before the draft model is loaded. If draft loading fails, place the draft model on CPU:

llama-server \
  -m /path/to/gemma-4-31B_q4_0-it.gguf \
  -md /path/to/mtp-gemma-4-31B-it.gguf \
  --spec-type draft-mtp \
  --spec-draft-ngl 0 \
  --spec-draft-n-max 4 \
  --kv-unified \
  -fa on

For example, with a 32 GB GPU and full 262k context, --spec-draft-ngl 0 may be needed. Reducing context length can allow the draft model to remain on GPU.

Notes

Requires a llama.cpp build with Gemma 4 MTP / draft-mtp support.
The draft model should be paired with the matching target model family and size.
These files are intended for speculative decoding acceleration, not direct prompting.
The 31B assistant checkpoint has n_ctx_train = 131072; using it with a 262k target may emit a context warning for the draft context.

License

The source Gemma 4 models are provided by Google under the Apache 2.0 license. See the original model cards and license terms:

Downloads last month: 562

GGUF

Model size

0.4B params

Architecture

gemma4-assistant

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for Pulsate1680/gemma-4-mtp-q8_0-gguf

Base model

google/gemma-4-12B-it-assistant

Finetuned

google/gemma-4-12B-it-qat-q4_0-unquantized-assistant

Quantized

(20)

this model