Qwen3.5-9B-abliterated-v2-MAX

Instructions to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Transformers:

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

pipe = pipeline("image-text-to-text", model="tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX")
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 AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX")
model = AutoModelForMultimodalLM.from_pretrained("tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX",
	filename="GGUF/Qwen3.5-9B-abliterated-v2-MAX.BF16.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"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX 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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
# Run inference directly in the terminal:
llama cli -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
# Run inference directly in the terminal:
llama cli -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX: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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
# Run inference directly in the terminal:
./llama-cli -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX: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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

Use Docker

docker model run hf.co/tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

LM Studio
Jan

vLLM

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX",
		"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/tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

SGLang

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX 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 "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX" \
    --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": "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX",
		"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 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 "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX" \
        --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": "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX",
		"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"
						}
					}
				]
			}
		]
	}'

Ollama
How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Ollama:
```
ollama run hf.co/tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
```

Unsloth Studio

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX 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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX 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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX to start chatting

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

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": "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

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 "tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16" \
  --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 tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Docker Model Runner:
```
docker model run hf.co/tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16
```

Lemonade

How to use tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX:BF16

Run and chat with the model

lemonade run user.Qwen3.5-9B-abliterated-v2-MAX-BF16

List all available models

lemonade list

Qwen3.5-9B-abliterated-v2-MAX

Qwen3.5-9B-abliterated-v2-MAX is an advanced unredacted evolution built on top of Qwen/Qwen3.5-9B. This version introduces a more optimized abliteration rate, combining refined refusal direction analysis with enhanced training strategies to further minimize internal refusal behaviors while preserving strong reasoning and instruction-following capabilities. The result is a highly capable 9B parameter language model designed for detailed responses and improved prompt adherence.

This model is intended strictly for research and learning purposes. Due to reduced internal refusal mechanisms, it may generate sensitive or unrestricted content. Users assume full responsibility for how the model is used. The authors and hosting platform disclaim any liability for generated outputs.

Compression for the Model

Qwen3.5-9B-abliterated-v2-MAX

Format	Description	Link
GGUF	Quantized GGUF format	https://huggingface.co/prithivMLmods/Qwen3.5-9B-abliterated-v2-MAX/tree/main/GGUF
NVFP4	NVFP4 compressed model	https://huggingface.co/prithivMLmods/Qwen3.5-9B-abliterated-v2-MAX-NVFP4
FP8	FP8 compressed model	https://huggingface.co/prithivMLmods/Qwen3.5-9B-abliterated-v2-MAX-FP8

Key Highlights

Optimized Abliteration Rate (v2): Improved suppression of refusal directions with better balance between openness and coherence.
Advanced Refusal Direction Analysis: Identifies and mitigates refusal-related activations within the model’s latent space.
Abliterated v2 Training Strategy: Further reduces refusal patterns while maintaining response quality and stability.
9B Parameter Architecture: Built on Qwen3.5-9B, offering strong reasoning while remaining efficient for modern GPUs.
Enhanced Instruction Adherence: Better handling of complex and nuanced prompts with minimal unnecessary refusals.
Efficient Deployment: Suitable for local inference, experimentation, and research workflows.

Quick Start with Transformers

pip install transformers==5.4.0
# or
pip install git+https://github.com/huggingface/transformers.git

from transformers import Qwen3_5ForConditionalGeneration, AutoProcessor
import torch

model = Qwen3_5ForConditionalGeneration.from_pretrained(
    "prithivMLmods/Qwen3.5-9B-abliterated-v2-MAX",
    torch_dtype="auto",
    device_map="auto"
)

processor = AutoProcessor.from_pretrained(
    "prithivMLmods/Qwen3.5-9B-abliterated-v2-MAX"
)

messages = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "Explain how transformer models work in simple terms."}
        ],
    }
]

text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)

inputs = processor(
    text=[text],
    padding=True,
    return_tensors="pt"
).to("cuda")

generated_ids = model.generate(**inputs, max_new_tokens=256)

generated_ids_trimmed = [
    out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]

output_text = processor.batch_decode(
    generated_ids_trimmed,
    skip_special_tokens=True,
    clean_up_tokenization_spaces=False
)

print(output_text)

Intended Use

Alignment & Refusal Research: Studying effects of aggressive abliteration and reduced refusal behavior.
Red-Teaming Experiments: Testing robustness across adversarial or edge-case prompts.
Local AI Deployment: Running high-capability models on consumer or high-end GPUs.
Research Prototyping: Exploring transformer behavior under modified alignment constraints.

Limitations & Risks

Important Note: This model intentionally minimizes built-in safety refusals.

High Risk of Sensitive Outputs: May generate unrestricted, controversial, or explicit content.
User Responsibility: Must be used within ethical, legal, and responsible boundaries.
Abliteration Trade-offs: Increased openness may occasionally reduce safety alignment or consistency.
Model Size Constraints: Despite improvements, a 9B model still has limits compared to larger frontier models.

Downloads last month: 256

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for tutuchen2000/Qwen3.5-9B-abliterated-v2-MAX

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Quantized

(345)

this model