Instructions to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Transformers:

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

pipe = pipeline("text-generation", model="morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored", dtype="auto")

llama-cpp-python

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored",
	filename="Qwen3.5-122B-A10B-Abliterix-v2-IQ4_NL.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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored 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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
# Run inference directly in the terminal:
llama cli -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
# Run inference directly in the terminal:
llama cli -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
# Run inference directly in the terminal:
./llama-cli -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

Use Docker

docker model run hf.co/morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

LM Studio
Jan

vLLM

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

SGLang

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored 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 "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored" \
    --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": "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored",
		"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 "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored" \
        --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": "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Ollama:
```
ollama run hf.co/morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
```

Unsloth Studio

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored 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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored 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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored to start chatting

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

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": "morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

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 morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Docker Model Runner:
```
docker model run hf.co/morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL
```

Lemonade

How to use morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull morikomorizz/Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored:IQ4_NL

Run and chat with the model

lemonade run user.Qwen3.5-122B-A10B-Abliterix-V2-Ultimate-Uncensored-IQ4_NL

List all available models

lemonade list

Qwen3.5-122B-A10B-Abliterix-Ultimate-Uncensored

This repository contains the GGUF quantized files for wangzhang/Qwen3.5-122B-A10B-abliterix.

Original Model: wangzhang/Qwen3.5-122B-A10B-abliterix
Architecture: Qwen3.5-122B-A10B
License: Apache 2.0
MTP Support: MTP Donor-unsloth/Qwen3.5-122B-A10B-MTP-GGUF

Quant Type	Size	Description
IQ4_NL	77 GB	Mixed Precision for Better Quality

Highlights

Metric	Value
Refusal rate	1/200 (0.5%)
KL divergence	0.0115
Optimization trials	25

The largest abliterated Qwen3.5 model. Only 1 out of 200 test prompts triggered a refusal — a 0.5% refusal rate with near-zero model degradation.

How It Works

Abliterix removes safety-refusal behavior while preserving model capabilities:

Refusal direction extraction — 800 harmful + 800 benign prompts reveal per-layer refusal activation patterns
Orthogonal projection — isolates the refusal signal by projecting out components aligned with normal responses, reducing refusals by 67% vs. raw abliteration
LoRA-based abliteration — rank-1 modifications to attention and MLP weights, captured as lightweight adapters (not destructive edits)
Bayesian optimization — Optuna TPE searches kernel shape, fractional direction index, and per-component strength across 25 trials to find the Pareto-optimal balance of low refusals and low KL divergence

Usage

llama.cpp

# IQ4_NL
./llama-cli -m Qwen3.5-122B-A10B-Abliterix-v2-IQ4_NL.gguf -p "Your prompt here" -n 512

llama-server

./llama-server -m Qwen3.5-122B-A10B-Abliterix-v2-IQ4_NL.gguf --host 0.0.0.0 --port 8080

Ollama

# Create a Modelfile
echo "FROM ./Qwen3.5-122B-A10B-Abliterix-v2-IQ4_NL.gguf" > Modelfile
ollama create qwen3.5-122b-abliterix -f Modelfile
ollama run qwen3.5-122b-abliterix

Citation

@software{abliterix,
  author = {Wu, Wangzhang},
  title = {Abliterix: Automated LLM Abliteration},
  year = {2026},
  url = {https://github.com/wuwangzhang1216/abliterix}
}