Instructions to use mlabonne/Qwen3-0.6B-abliterated-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="mlabonne/Qwen3-0.6B-abliterated-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mlabonne/Qwen3-0.6B-abliterated-GGUF", dtype="auto")

llama-cpp-python

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mlabonne/Qwen3-0.6B-abliterated-GGUF",
	filename="qwen3-0.6b-abliterated.q2_k.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mlabonne/Qwen3-0.6B-abliterated-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 mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mlabonne/Qwen3-0.6B-abliterated-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 mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mlabonne/Qwen3-0.6B-abliterated-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 mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlabonne/Qwen3-0.6B-abliterated-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": "mlabonne/Qwen3-0.6B-abliterated-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

SGLang

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "mlabonne/Qwen3-0.6B-abliterated-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/Qwen3-0.6B-abliterated-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "mlabonne/Qwen3-0.6B-abliterated-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/Qwen3-0.6B-abliterated-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Ollama:
```
ollama run hf.co/mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
```

Unsloth Studio

How to use mlabonne/Qwen3-0.6B-abliterated-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 mlabonne/Qwen3-0.6B-abliterated-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 mlabonne/Qwen3-0.6B-abliterated-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mlabonne/Qwen3-0.6B-abliterated-GGUF to start chatting

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

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": "mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

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 mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Docker Model Runner:
```
docker model run hf.co/mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M
```

Lemonade

How to use mlabonne/Qwen3-0.6B-abliterated-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mlabonne/Qwen3-0.6B-abliterated-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-0.6B-abliterated-GGUF-Q4_K_M

List all available models

lemonade list

🐹 Qwen3-0.6B-abliterated

Qwen3 Abliterated 0.6B • 1.7B • 4B • 8B • 14B • 30B-A3B

This is an uncensored version of Qwen/Qwen3-0.6B created with a new abliteration technique. See this article to know more about abliteration.

This is a research project to understand how refusals and latent fine-tuning work in LLMs. I played with different sizes of Qwen3 and noticed there was no one-size-fits-all abliteration strategy. In addition, the reasoning mode interfered with non-reasoning refusals, which made it more challenging. This made me iterate over different recipes and significantly consolidate my scripts with accumulation and better evaluations.

Note that this is fairly experimental, so it might not turn out as well as expected.

I recommend using these generation parameters: temperature=0.6, top_k=20, top_p=0.95, min_p=0.

✂️ Abliteration

The refusal direction is computed by comparing the residual streams between target (harmful) and baseline (harmless) samples. The hidden states of target modules (e.g., o_proj) are orthogonalized to subtract this refusal direction with a given weight factor. These weight factors follow a normal distribution with a certain spread and peak layer. Modules can be iteratively orthogonalized in batches, or the refusal direction can be accumulated to save memory.

Finally, I used a hybrid evaluation with a dedicated test set to calculate the acceptance rate. This uses both a dictionary approach and NousResearch/Minos-v1. The goal is to obtain an acceptance rate >90% and still produce coherent outputs.

Downloads last month: 240

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Model tree for mlabonne/Qwen3-0.6B-abliterated-GGUF

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Quantized

(324)

this model

Collection including mlabonne/Qwen3-0.6B-abliterated-GGUF

✂️ Abliteration

Collection

Uncensored models using abliteration. See this article for more information: huggingface.co/blog/mlabonne/abliteration • 32 items • Updated Mar 2 • 167