Instructions to use matrixportalx/Gemma-3-4B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use matrixportalx/Gemma-3-4B-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("matrixportalx/Gemma-3-4B-GGUF", dtype="auto")

llama-cpp-python

How to use matrixportalx/Gemma-3-4B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="matrixportalx/Gemma-3-4B-GGUF",
	filename="gemma-3-4b.q4_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 matrixportalx/Gemma-3-4B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf matrixportalx/Gemma-3-4B-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 matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf matrixportalx/Gemma-3-4B-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 matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf matrixportalx/Gemma-3-4B-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 matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf matrixportalx/Gemma-3-4B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/matrixportalx/Gemma-3-4B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use matrixportalx/Gemma-3-4B-GGUF with Ollama:
```
ollama run hf.co/matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
```

Unsloth Studio

How to use matrixportalx/Gemma-3-4B-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 matrixportalx/Gemma-3-4B-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 matrixportalx/Gemma-3-4B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for matrixportalx/Gemma-3-4B-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use matrixportalx/Gemma-3-4B-GGUF with Docker Model Runner:
```
docker model run hf.co/matrixportalx/Gemma-3-4B-GGUF:Q4_K_M
```

Lemonade

How to use matrixportalx/Gemma-3-4B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull matrixportalx/Gemma-3-4B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Gemma-3-4B-GGUF-Q4_K_M

List all available models

lemonade list

Matrix commited on Jul 10, 2025

Commit

b0455d6

verified ·

1 Parent(s): 941f6df

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +33 -0

README.md ADDED Viewed

	@@ -0,0 +1,33 @@

+---
+base_model:
+- google/gemma-3-4b-it-qat-int4-unquantized
+- huihui-ai/gemma-3-4b-it-abliterated
+- VIDraft/Gemma-3-R1984-4B
+- neo4j/text-to-cypher-Gemma-3-4B-Instruct-2025.04.0
+- google/medgemma-4b-it
+- google/gemma-3-4b-it-qat-q4_0-unquantized
+- ZySec-AI/gemma-3-4b-document-writer
+library_name: transformers
+tags:
+- mergekit
+- merge
+---
+# Gemma-3-4B GGUF Quantized Models
+## Technical Details
+- **Quantization Tool:** llama.cpp
+- **Version:** version: 5862 (704bb7a7)
+## Model Information
+- **Base Model:** [Gunulhona/Gemma-3-4B](https://huggingface.co/Gunulhona/Gemma-3-4B)
+- **Quantized by:** [matrixportal](https://huggingface.co/matrixportal)
+## Available Files
+| 🚀 Download | 🔢 Type | 📝 Description |
+|------------|---------|---------------|
+| [Download](https://huggingface.co/matrixportal/Gemma-3-4B-GGUF/resolve/main/gemma-3-4b.q4_0.gguf) | Q4 0 | Standard 4-bit (fast on ARM) |
+| [Download](https://huggingface.co/matrixportal/Gemma-3-4B-GGUF/resolve/main/gemma-3-4b.q4_k_m.gguf) | Q4 K M | 4-bit balanced (recommended default) |
+| [Download](https://huggingface.co/matrixportal/Gemma-3-4B-GGUF/resolve/main/gemma-3-4b.q5_k_m.gguf) | Q5 K M | 5-bit best (recommended HQ option) |
+💡 **Q4 K M** provides the best balance for most use cases