Instructions to use SandLogicTechnologies/Gemma-3-270m-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SandLogicTechnologies/Gemma-3-270m-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SandLogicTechnologies/Gemma-3-270m-GGUF",
	filename="gemma-3-270m_Q4_k_m.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use SandLogicTechnologies/Gemma-3-270m-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Gemma-3-270m-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 SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Gemma-3-270m-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 SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf SandLogicTechnologies/Gemma-3-270m-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 SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M

Use Docker

docker model run hf.co/SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use SandLogicTechnologies/Gemma-3-270m-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SandLogicTechnologies/Gemma-3-270m-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": "SandLogicTechnologies/Gemma-3-270m-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M

Ollama
How to use SandLogicTechnologies/Gemma-3-270m-GGUF with Ollama:
```
ollama run hf.co/SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
```

Unsloth Studio new

How to use SandLogicTechnologies/Gemma-3-270m-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 SandLogicTechnologies/Gemma-3-270m-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 SandLogicTechnologies/Gemma-3-270m-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SandLogicTechnologies/Gemma-3-270m-GGUF to start chatting

Docker Model Runner
How to use SandLogicTechnologies/Gemma-3-270m-GGUF with Docker Model Runner:
```
docker model run hf.co/SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M
```

Lemonade

How to use SandLogicTechnologies/Gemma-3-270m-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SandLogicTechnologies/Gemma-3-270m-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Gemma-3-270m-GGUF-Q4_K_M

List all available models

lemonade list

Quantized Google Gemma-3-270M Model

This repository provides the Gemma-3-270M model, one of Google’s lightweight and efficient open models from the Gemma family. With only 270 million parameters, it is designed for fast inference, prototyping, and edge deployment, while retaining strong general-purpose reasoning, text understanding, and instruction-following abilities.

Model Overview

Original Model: Gemma-3-270M
Architecture: Decoder-only Transformer
Base Model: Gemma-3 series
Modalities: Text only
Developer: Google DeepMind
License: Gemma Terms of Use
Language: English

Quantization Details

Q4_K_M Version

Approx. ~54% size reduction
Lower memory footprint (~241 MB)
Slight performance degradation in complex reasoning scenarios

Q5_K_M Version

Approx. ~52% size reduction
Higher fidelity (~247 MB)
Better performance retention, recommended when quality is a priority

Key Features

Extremely lightweight (270M parameters) – easy to run on CPU, edge devices, or low-resource GPUs
General-purpose instruction following
Useful for text completion, reasoning, and lightweight assistant prototypes
Designed to be fast and memory efficient while maintaining quality
Good foundation for research, experimentation, and fine-tuning

llama.cpp (text-only)

./llama-cli -hf SandLogicTechnologies/gemma-3-270m_Q4_k_m -p "What is LLM Quantization"

Usage

This model is intended for developers and researchers who need a lightweight LLM for prototyping and small-scale applications.

Acknowledgments

These quantized models are based on the original work by Google development team.

Special thanks to:

The Google team for developing and releasing the gemma-3-270m model.
Georgi Gerganov and the entire llama.cpp open-source community for enabling efficient model quantization and inference via the GGUF format.

Contact

For any inquiries or support, please contact us at support@sandlogic.com or visit our Website.

Downloads last month: 25

GGUF

Model size

0.3B params

Architecture

gemma3

Hardware compatibility

4-bit

5-bit

Model tree for SandLogicTechnologies/Gemma-3-270m-GGUF

Base model

google/gemma-3-270m

Quantized

(37)

this model