Instructions to use c516a/gemma-3-12b-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use c516a/gemma-3-12b-it with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="c516a/gemma-3-12b-it",
	filename="gemma-3-12b-it.Q2_K.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 c516a/gemma-3-12b-it with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf c516a/gemma-3-12b-it:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf c516a/gemma-3-12b-it:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf c516a/gemma-3-12b-it:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf c516a/gemma-3-12b-it: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 c516a/gemma-3-12b-it:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf c516a/gemma-3-12b-it: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 c516a/gemma-3-12b-it:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf c516a/gemma-3-12b-it:Q4_K_M

Use Docker

docker model run hf.co/c516a/gemma-3-12b-it:Q4_K_M

LM Studio
Jan

vLLM

How to use c516a/gemma-3-12b-it with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "c516a/gemma-3-12b-it"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c516a/gemma-3-12b-it",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/c516a/gemma-3-12b-it:Q4_K_M

Ollama
How to use c516a/gemma-3-12b-it with Ollama:
```
ollama run hf.co/c516a/gemma-3-12b-it:Q4_K_M
```

Unsloth Studio new

How to use c516a/gemma-3-12b-it 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 c516a/gemma-3-12b-it 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 c516a/gemma-3-12b-it to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for c516a/gemma-3-12b-it to start chatting

Docker Model Runner
How to use c516a/gemma-3-12b-it with Docker Model Runner:
```
docker model run hf.co/c516a/gemma-3-12b-it:Q4_K_M
```

Lemonade

How to use c516a/gemma-3-12b-it with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull c516a/gemma-3-12b-it:Q4_K_M

Run and chat with the model

lemonade run user.gemma-3-12b-it-Q4_K_M

List all available models

lemonade list

⚖️ License and Usage

This repository contains quantized variants of the Gemma language model developed by Google.
The original models and their license can be found at: https://ai.google.dev/gemma/terms

🧠 Model source: Google / Gemma
🪄 Quantized by: c516a

These quantized models are:

Provided under the same terms as the original Google Gemma models.
Intended only for non-commercial use, research, and experimentation.
Redistributed without modification to the underlying model weights, except for format (GGUF) and quantization level.

By using this repository or its contents, you agree to:

Comply with the Gemma License Terms,
Not use the model or its derivatives for any commercial purposes without a separate license from Google,
Acknowledge Google as the original model creator.

📢 This repository is not affiliated with Google.

Downloads last month: 27

GGUF

Model size

12B params

Architecture

gemma3

Hardware compatibility

2-bit

3-bit

4-bit

6-bit

View +1 variant

Model tree for c516a/gemma-3-12b-it

Base model

google/gemma-3-12b-pt

Finetuned

google/gemma-3-12b-it

Quantized

(151)

this model