ikarius/Qwen3-14B-Abliterated-NF4

Qwen3-14B-Abliterated-v2-nf4

Model Overview

This repository contains a quantized version (NF4, using BitsAndBytes) of the Qwen3-14B-Abliterated-v2 model.
The original model is an uncensored variant of Qwen/Qwen3-14B, created using the abliteration technique to remove refusal behaviors (see remove-refusals-with-transformers).

This quantization was performed by ikarius to reduce model size and enable efficient inference on consumer hardware, while preserving the uncensored capabilities of the base model.

Key Features:

• Base Model: Qwen3-14B (abliterated for uncensoring)
• Version: Abliterated-v2 (improved over v1)
• Quantization: NF4 (4-bit NormalFloat via BitsAndBytes)
• Parameters: 14 Billion
• License: Refer to the original Qwen3 license (Apache 2.0 with additional terms); abliteration does not alter the license.
• Intended Use: Research, experimentation, and creative applications.

  Warning: This model is uncensored and may generate sensitive or harmful content—use responsibly.

---

Installation

1. Install the required dependencies:

   pip install transformers torch bitsandbytes accelerate
   

Ensure you have a compatible CUDA setup for GPU acceleration.

(Optional) For CPU-only inference:bash

pip install optimum[exporters]

Usage Load and run the model using Hugging Face Transformers. Python Example

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig import torch

NF4 quantization config

quantization_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, )

Load tokenizer and model

tokenizer = AutoTokenizer.from_pretrained(MODEL_ID) model = AutoModelForCausalLM.from_pretrained( MODEL_ID, quantization_config=quantization_config, device_map="auto", trust_remote_code=True, )

Example inference

prompt = "Hello, how are you?" inputs = tokenizer(prompt, return_tensors="pt").to(model.device) outputs = model.generate( **inputs, max_new_tokens=128, temperature=0.7, do_sample=True, pad_token_id=tokenizer.eos_token_id, ) response = tokenizer.decode(outputs[0], skip_special_tokens=True) print(response)

Inference Tips

VRAM: ~8–10 GB required for 14B NF4 on a single GPU. Batch Size: Start with 1. Thinking Mode: v2 supports step-by-step reasoning prompts. Streaming: Use TextStreamer for real-time output.

Quantization Details

Method: BitsAndBytes NF4 (normal float 4-bit) Quantizer: ikarius Benefits: ~75% size reduction vs FP16, minimal quality loss Trade-offs: Slight perplexity increase

Reproduce Quantization

from transformers import AutoModelForCausalLM, BitsAndBytesConfig import torch

original_model = AutoModelForCausalLM.from_pretrained( "huihui-ai/Huihui-Qwen3-14B-abliterated-v2", torch_dtype=torch.float16, device_map="auto", )

quant_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, )

quantized_model = AutoModelForCausalLM.from_pretrained( "huihui-ai/Huihui-Qwen3-14B-abliterated-v2", quantization_config=quant_config, device_map="auto", )

quantized_model.save_pretrained("Qwen3-14B-Abliterated-v2-nf4") tokenizer.save_pretrained("Qwen3-14B-Abliterated-v2-nf4")

Limitations & Ethics

May amplify training data biases. Not suitable for production without alignment. For commercial use: review original licenses.

Contact

Open an issue or reach out to ikarius on Hugging Face.

Last updated: November 12, 2025

Original Model Credits-

Abliteration by:huihui-ai →Support huihui-ai: Buy huihui-ai a coffee ☕ Base Model:Qwen/Qwen3-14B