Instructions to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF",
	filename="gemma-4-31B-it-heretic-QAT-Q2_K_L.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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF 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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
# Run inference directly in the terminal:
llama cli -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
# Run inference directly in the terminal:
llama cli -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
# Run inference directly in the terminal:
./llama-cli -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

Use Docker

docker model run hf.co/SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

LM Studio
Jan
Ollama
How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with Ollama:
```
ollama run hf.co/SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
```

Unsloth Studio

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF to start chatting

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

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": "SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

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 SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with Docker Model Runner:
```
docker model run hf.co/SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L
```

Lemonade

How to use SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SC117/gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF:Q2_K_L

Run and chat with the model

lemonade run user.gemma-4-31B-it-heretic-QAT-For-Edge-16G-GGUF-Q2_K_L

List all available models

lemonade list

⚡ Gemma 4 31B Heretic QAT — Extreme Quantization for 16GB VRAM

Heretic QAT · Q2_K + Q8_0 Embedding · 11.4 GB · Fits 16GB VRAM

📖 中文文档

Q2_K_L 31B Dense Heretic Uncensored 11.4 GB QAT Weights ⚠️ 16GB VRAM

Extreme quantization of coder3101/heretic-QAT using Unsloth's Q2_K_L recipe (Q2_K body + Q8_0 embedding/output). Compresses the 31B dense model to 11.4 GB, enabling it to run on 16GB VRAM consumer GPUs.

⚠️ Extreme Quantization Warning — Please Read

This quantization uses Q2_K (~2.9 BPW body) to compress the 31B model to 11.4 GB. This is an extreme solution for running 31B on 16GB VRAM, with known limitations:

Repetition loops — Model may get stuck outputting the same token or phrase repeatedly
Inaccurate arithmetic — Multi-digit math calculations are unreliable
Degraded complex reasoning — Multi-step logic, math proofs significantly worse than higher precision
Possible hallucinations — Factual QA accuracy is reduced

Suitable for (low-stakes tasks):

Creative writing, story generation, brainstorming
Casual conversation, chat, roleplay
Code assistance (simple functions, code explanation, comment generation)
Text summarization, translation, format conversion
Non-precise general knowledge Q&A, concept explanation

Not recommended for: Math, precise reasoning, fact-checking, production use, or any task requiring strict output accuracy.

✂️ Heretic ARA Abliteration Parameters

Base: coder3101/gemma-4-31B-it-qat-q4_0-unquantized-heretic · Heretic v1.2.0 · ARA + Row-Norm

Parameter	Value
start_layer_index	15
end_layer_index	39
preserve_good_behavior_weight	0.6649
steer_bad_behavior_weight	0.0001
overcorrect_relative_weight	1.0818
neighbor_count	15

Metric	Heretic	Original QAT
KL Divergence	0.0564	0 (by definition)
Refusals	11/100	99/100

🏗️ Architecture

Base Model	google/gemma-4-31B-it
Parameters	30.7B (dense, all parameters active)
Architecture	Hybrid: Gated DeltaNet linear attention + Softmax global attention (3:1 alternating)
Layers	60
Context Length	256K tokens (supported by this quantization)
Vocabulary	262K, 140+ languages
Modalities	Text + Image (native multimodal)
QAT Training	Google official QAT (quantization-aware), weights inherently robust to quantization
Quantization	Q2_K_L (Q2_K body + Q8_0 embedding/output), b9553 llama-quantize

⚙️ Recommended Sampling Parameters

Thinking mode (default):

General	temp=1.0, top_p=0.95, top_k=64
Coding	temp=0.6, top_p=0.95, top_k=64

Use --jinja flag with llama.cpp. Disable thinking: --chat-template-kwargs '{"enable_thinking":false}'.

📝 Usage

Compatible with llama.cpp, LM Studio, Jan, koboldcpp, and other GGUF runtimes. Use --jinja flag and -ngl 99 for GPU offload.

llama-cli -m gemma-4-31B-it-heretic-QAT-Q2_K_L.gguf \
  --jinja -ngl 99 -c 4096 \
  --repeat-penalty 1.3

📊 Quantization Details

Format	Q2_K_L (Unsloth recipe: Q2_K + Q8_0 embedding + Q8_0 output)
File Size	11.4 GB
Effective BPW	~3.16 (body ~2.9 + embedding/output ~8.0)
Tool	llama-quantize (b9553, CUDA 13.3)
Source	F16 GGUF (converted from QAT heretic safetensors)
Context Length	256K (set in GGUF metadata)
QAT Advantage	QAT weights significantly outperform regular BF16 at Q4_0; advantage diminishes but persists at Q2 level

🔗 Credits

Heretic Abliteration: coder3101 · Heretic v1.2.0 ARA + Row-Norm
QAT Weights: Google Gemma 4 QAT
Quantization Recipe: Unsloth Q2_K_L (Q2_K + Q8_0 embedding/output)
Quantization Tool: llama.cpp b9553 · GitHub
Original Model: Google Gemma 4 31B IT