ImageShield-MMCF-0.8B

ImageShield-MMCF — Multimodal Content Filter is a multimodal content-safety classifier built on top of Qwen/Qwen3.5-0.8B and trained on approximately 28,000 content-safety guardrail samples. The model is designed to classify visual content as Safe or Unsafe, with a particular focus on detecting Non-Consensual Intimate Imagery (NCII) and other potentially sensitive visual content.

This model is an experimental release and may generate unexpected behaviors or reasoning artifacts in certain scenarios.

This model is intended for research and content safety classification only. Any outputs generated by this model are the sole responsibility of the user. The authors and hosting platform disclaim all liability for generated content. Users must ensure safe, ethical, and lawful usage.

Key Highlights

• Qwen 3.5 Foundation: Built on top of Qwen/Qwen3.5-0.8B.
• Multimodal Content Filter: Designed for visual content-safety classification.
• 28K Training Samples: Trained on approximately 28,000 content-safety guardrail samples.
• Safe / Unsafe Classification: Classifies visual content as either Safe or Unsafe.
• NCII Detection: Focused on identifying potentially Non-Consensual Intimate Imagery (NCII) and related sensitive visual content.
• Zero-Tolerance Safety Classification: Uses a conservative classification policy that favors safety when the content is ambiguous.
• Structured Output: Produces a machine-readable JSON response containing a caption, classification, and reasoning.

Standard Instruction

You are a strict, zero-tolerance parental control content-safety guardrail classifier for visual media.

Your role is minor protection: any media containing suggestive, revealing, sexually provocative, or explicit visual elements must be flagged as NSFW.

STRICT ZERO-TOLERANCE CLASSIFICATION RULES:

1. Nudity & Exposure: Any full or partial nudity, exposed breasts, cleavage, buttocks, groin, or visible pubic area MUST be classified as NSFW.

2. Underwear & Swimwear: Thongs, G-strings, micro-bikinis, skimpy or revealing swimwear, bras, panties, lingerie, see-through clothing, or unbuttoned tops revealing undergarments or bare torsos MUST be classified as NSFW.

3. Poses & Suggestiveness: Sexually suggestive poses, seductive modeling, cleavage-emphasizing framing, fetish content, or explicit erotic themes MUST be classified as NSFW.

4. Artwork & Animation: Anime nudes, ecchi, hentai, suggestive illustrations, or 2D/3D stylized erotica MUST be classified as NSFW.

5. Classification Threshold: When in doubt, err on the side of caution and classify the content as NSFW.

Output strictly valid JSON with no extra conversational text, commentary, or markdown formatting outside the JSON object:

{
  "caption": "<Detailed, objective visual description of the subject, clothing, exposure, and pose>",
  "is_nsfw": true | false,
  "reason": "<Precise reason for classification based on clothing, exposure, or pose>",
  "nsfw": 1 | 0,
  "safe": 1 | 0
}

Quick Start with Transformers

pip install transformers
pip install accelerate

from transformers import Qwen3_5ForConditionalGeneration, AutoProcessor
import torch

model = Qwen3_5ForConditionalGeneration.from_pretrained(
    "prithivMLmods/ImageShield-MMCF-0.8B",
    torch_dtype="auto",
    device_map="auto"
)

processor = AutoProcessor.from_pretrained(
    "prithivMLmods/ImageShield-MMCF-0.8B"
)

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": "Classify this image using the provided content-safety guardrail."
            }
        ],
    }
]

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
)

output_text = processor.batch_decode(
    [
        out[len(inp):]
        for inp, out in zip(inputs.input_ids, generated_ids)
    ],
    skip_special_tokens=True,
    clean_up_tokenization_spaces=False
)

print(output_text[0])

---

Example Cards

{
  "caption": "A woman in a white tank top stands outdoors, gesturing with her right hand near her face while speaking. She wears a silver necklace and a bracelet. To her right, a man in a dark shirt is partially visible, looking towards her. The background shows blurred pedestrians and urban buildings.",
  "is_nsfw": false,
  "reason": "The image depicts a casual outdoor scene featuring two adults in everyday attire. There is no evidence of nudity, revealing undergarments, suggestive poses, or any sexually explicit content. The focus is on the interaction between the subjects.",
  "nsfw": 0,
  "safe": 1
}

---

{
  "caption": "A blonde woman is shown from behind, wearing red thong-style underwear. She has long hair and a visible tattoo on her lower back. Her hands are positioned near her hips, holding the straps of the thong. The setting appears to be a modern living room with a window and a patterned sofa.",
  "is_nsfw": 1,
  "reason": "The image classifies as NSFW due to the presence of red thong-style underwear, which exposes the lower back and buttocks. The pose, where the subject is holding the thong straps, combined with the visible skin and lack of full coverage, meets the criteria for suggestive and revealing content.",
  "nsfw": 1,
  "safe": 0
}

---

{
  "caption": "A medium-sized lioness with golden-brown fur walks across a grassy enclosure in a zoo setting. She is walking towards the camera, looking directly at the viewer. The background features large rocks, trees, and a thatched-roof structure, likely a viewing platform or shelter. The environment appears clean and well-maintained.",
  "is_nsfw": false,
  "reason": "The image depicts a standard outdoor scene featuring a lioness in a zoo environment. There is no evidence of nudity, revealing clothing, suggestive poses, or any sexually explicit content. The focus is entirely on the animal and its natural habitat.",
  "nsfw": 0,
  "safe": 1
}

---

Training Details

Setting	Value
Base Model	Qwen/Qwen3.5-0.8B
Model Type	Multimodal Content-Safety Classifier
Training Samples	Approximately 28,000
Training Objective	Content-safety guardrail classification
Primary Classification	Safe / Unsafe
Safety Focus	Non-Consensual Intimate Imagery (NCII)
Training Framework	TRL

Intended Use

• Content Safety Classification: Classifying visual media as safe or unsafe.
• NCII Detection: Supporting research into automated detection of potentially non-consensual intimate imagery.
• Parental Controls: Building conservative visual content-safety filtering systems.
• Content Moderation: Supporting automated safety classification pipelines.
• Multimodal Safety Research: Evaluating content-safety behavior in multimodal language models.
• Guardrail Development: Researching structured safety classification and filtering workflows.

Limitations

• Experimental Model: The model may produce incorrect or inconsistent classifications.
• False Positives: Benign content may occasionally be classified as unsafe due to the conservative classification threshold.
• False Negatives: Unsafe content may occasionally be missed.
• Context Sensitivity: Classification performance depends on image quality, visual context, and the provided instruction.
• Automated Classification: The model should not be treated as a definitive legal or safety determination.

Acknowledgements

• Qwen/Qwen3.5-0.8B: Base multimodal model used for this project.

• TRL – Transformers Reinforcement Learning: TRL is a full-stack library providing tools to train transformer language models with methods including Supervised Fine-Tuning (SFT), Group Relative Policy Optimization (GRPO), Direct Preference Optimization (DPO), Reward Modeling, and more.

• Transformers: Transformers provides state-of-the-art machine learning models for text, computer vision, audio, video, and multimodal tasks, supporting both inference and training.