license: apache-2.0
pipeline_tag: image-text-to-text
library_name: transformers
base_model:

• Qwen/Qwen3-VL-30B-A3B-Thinking
  tags:
• abliterated
• uncensored
• trl

huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated

This is the uncensored version of Qwen/Qwen3-VL-30B-A3B-Thinking after fine-tuning. It has been trained using TRL.

It was only the text part that was processed, not the image part.

The abliterated model will no longer say "I can’t describe or analyze this image."

Framework versions

• TRL: 0.24.0
• Transformers: 4.57.1
• Pytorch: 2.8.0+cu128
• Datasets: 4.0.0
• Tokenizers: 0.22.1

TrainingMetrics

The fine-tuning is still ongoing. We will test the subsequent results of the fine-tuning, and if feasible,
we might release a new version.

ollama

Please update to the latest version of Ollama-v0.12.7.
**New Update
You can use huihui_ai/qwen3-vl-abliterated:30b-a3b-Thinking-v2 directly,

After multiple tests, a version was finally selected in step 2050. For specific details, please refer to 2050.

ollama run huihui_ai/qwen3-vl-abliterated:30b-a3b-Thinking-v2

old

You can use huihui_ai/qwen3-vl-abliterated:30b-a3b-thinking directly,

ollama run huihui_ai/qwen3-vl-abliterated:30b

Chat with Image

from transformers import Qwen3VLMoeForConditionalGeneration, AutoProcessor, BitsAndBytesConfig
import os
import torch

cpu_count = os.cpu_count()
print(f"Number of CPU cores in the system: {cpu_count}")
half_cpu_count = cpu_count // 2
os.environ["MKL_NUM_THREADS"] = str(half_cpu_count)
os.environ["OMP_NUM_THREADS"] = str(half_cpu_count)
torch.set_num_threads(half_cpu_count)

MODEL_ID = "huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated"

# default: Load the model on the available device(s)
model = Qwen3VLMoeForConditionalGeneration.from_pretrained(
    MODEL_ID, 
    device_map="auto", 
    trust_remote_code=True,
    dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
)
# We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.
# model = Qwen3VLMoeForConditionalGeneration.from_pretrained(
#     "Qwen/Qwen3-VL-235B-A22B-Instruct",
#     dtype=torch.bfloat16,
#     attn_implementation="flash_attention_2",
#     device_map="auto",
# )

processor = AutoProcessor.from_pretrained(MODEL_ID)


image_path = "/png/cars.jpg"

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image", "image": f"{image_path}",
            },
            {"type": "text", "text": "Describe this image."},
        ],
    }
]

# Preparation for inference
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt"
).to(model.device)

# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=128)
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)

Usage Warnings

• Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.

• Not Suitable for All Audiences: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.

• Legal and Ethical Responsibilities: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.

• Research and Experimental Use: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.

• Monitoring and Review Recommendations: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.

• No Default Safety Guarantees: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use.

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.

• bitcoin:

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

• Support our work on Ko-fi!