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

Libraries

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

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Sabomako/gemma-3-12b-it-heretic",
	filename="gguf/gemma3-12b-it-heretic-Q8.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": [
				{
					"type": "text",
					"text": "Describe this image in one sentence."
				},
				{
					"type": "image_url",
					"image_url": {
						"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
					}
				}
			]
		}
	]
)

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Sabomako/gemma-3-12b-it-heretic 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 Sabomako/gemma-3-12b-it-heretic:BF16
# Run inference directly in the terminal:
llama cli -hf Sabomako/gemma-3-12b-it-heretic:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Sabomako/gemma-3-12b-it-heretic:BF16
# Run inference directly in the terminal:
llama cli -hf Sabomako/gemma-3-12b-it-heretic:BF16

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 Sabomako/gemma-3-12b-it-heretic:BF16
# Run inference directly in the terminal:
./llama-cli -hf Sabomako/gemma-3-12b-it-heretic:BF16

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 Sabomako/gemma-3-12b-it-heretic:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Sabomako/gemma-3-12b-it-heretic:BF16

Use Docker

docker model run hf.co/Sabomako/gemma-3-12b-it-heretic:BF16

LM Studio
Jan

vLLM

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

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Sabomako/gemma-3-12b-it-heretic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Sabomako/gemma-3-12b-it-heretic",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Sabomako/gemma-3-12b-it-heretic:BF16

Ollama
How to use Sabomako/gemma-3-12b-it-heretic with Ollama:
```
ollama run hf.co/Sabomako/gemma-3-12b-it-heretic:BF16
```

Unsloth Studio

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

Using HuggingFace Spaces for Unsloth

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

Atomic Chat new
Docker Model Runner
How to use Sabomako/gemma-3-12b-it-heretic with Docker Model Runner:
```
docker model run hf.co/Sabomako/gemma-3-12b-it-heretic:BF16
```

Lemonade

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

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Sabomako/gemma-3-12b-it-heretic:BF16

Run and chat with the model

lemonade run user.gemma-3-12b-it-heretic-BF16

List all available models

lemonade list

This is a decensored version of google/gemma-3-12b-it, made using Heretic v1.2.0 with (MPOA) Magnitude-Preserving Orthogonal Ablation

Abliteration parameters

Parameter	Value
direction_index	20.58
attn.o_proj.max_weight	1.29
attn.o_proj.max_weight_position	28.86
attn.o_proj.min_weight	0.98
attn.o_proj.min_weight_distance	23.38
mlp.down_proj.max_weight	1.36
mlp.down_proj.max_weight_position	29.82
mlp.down_proj.min_weight	0.77
mlp.down_proj.min_weight_distance	25.21

Performance

Metric	This model	Original model (google/gemma-3-12b-it)
KL divergence	0.024	0 (by definition)
Refusals	4/100	97/100

Downloads last month: 72

Safetensors

Model size

12B params

Tensor type

BF16

Model tree for Sabomako/gemma-3-12b-it-heretic

Base model

google/gemma-3-12b-pt

Finetuned

google/gemma-3-12b-it

Quantized

(164)

this model