GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

prithivMLmods/OneDecision-VisionGuard-4B-SFT-GGUF overview

OneDecision VisionGuard 4B SFT GGUF assets/banner.gif <p align="center" <a href="https://huggingface.co/prithivMLmods/OneDecision VisionGuard 27B SFT model var…

transformersggufdecision-modeltext-generation-inferencellama-cppSFTimage-classificationOneDecisionVisionGuardJSONcontent-safetycontent-moderationmultimodalsafety-classifierguardrailvisual-safetymultimodal-content-filteruncensoredimage-text-to-textendataset:prithivMLmods/ImageShield-OneDecision-Classificationbase_model:prithivMLmods/OneDecision-VisionGuard-4B-SFTbase_model:quantized:prithivMLmods/OneDecision-VisionGuard-4B-SFTlicense:apache-2.0

Runs locally from ~644.3 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
378
Likes
1
Pipeline
image-text-to-text

Repository Files & Downloads

9 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
OneDecision-VisionGuard-4B-SFT.BF16.ggufGGUFGGUF7.85 GBDownload
OneDecision-VisionGuard-4B-SFT.Q3_K_L.ggufGGUFGGUF2.26 GBDownload
OneDecision-VisionGuard-4B-SFT.Q3_K_M.ggufGGUFGGUF2.11 GBDownload
OneDecision-VisionGuard-4B-SFT.Q4_K_M.ggufGGUFGGUF2.52 GBDownload
OneDecision-VisionGuard-4B-SFT.Q4_K_S.ggufGGUFGGUF2.39 GBDownload
OneDecision-VisionGuard-4B-SFT.Q5_K_M.ggufGGUFGGUF2.86 GBDownload
OneDecision-VisionGuard-4B-SFT.Q5_K_S.ggufGGUFGGUF2.78 GBDownload
OneDecision-VisionGuard-4B-SFT.Q6_K.ggufGGUFGGUF3.23 GBDownload
OneDecision-VisionGuard-4B-SFT.mmproj-bf16.ggufGGUFBF16644.3 MBDownload

Model Details

Model IDprithivMLmods/OneDecision-VisionGuard-4B-SFT-GGUF
AuthorprithivMLmods
Pipelineimage-text-to-text
Licenseapache-2.0
Base modelprithivMLmods/OneDecision-VisionGuard-4B-SFT
Last modified2026-10-07T15:29:17.000Z

Model README

---

tags:

  • decision-model
  • text-generation-inference
  • llama-cpp
  • SFT
  • image-classification
  • OneDecision
  • VisionGuard
  • JSON
  • content-safety
  • content-moderation
  • multimodal
  • safety-classifier
  • guardrail
  • visual-safety
  • multimodal-content-filter
  • uncensored

datasets:

  • prithivMLmods/ImageShield-OneDecision-Classification

license: apache-2.0

language:

  • en

base_model:

  • prithivMLmods/OneDecision-VisionGuard-4B-SFT

pipeline_tag: image-text-to-text

library_name: transformers

---

!OneDecision-VisionGuard-4B-SFT-GGUF

<p align="center">

<a href="https://huggingface.co/prithivMLmods/OneDecision-VisionGuard-27B-SFT#model-variants" target="_blank">📦 Model Variants</a> |

<a href="https://huggingface.co/spaces/prithivMLmods/OneDecision-VisionGuard-Demo" target="_blank">🤗 HF Demo</a> |

<a href="https://huggingface.co/collections/prithivMLmods/onedecision-visionguard" target="_blank">🛢️ Collection</a> |

<a href="https://github.com/PRITHIVSAKTHIUR/OneDecision-VisionGuard-Demo" target="_blank">🔗 GitHub</a>

</p>

OneDecision-VisionGuard-4B-SFT-GGUF

> OneDecision-VisionGuard-4B-SFT is a dense 4-billion-parameter multimodal image classification model based on Qwen/Qwen3.5-4B and trained on the ImageShield-OneDecision-Classification content-safety guardrail dataset. The model is designed to classify visual content as Safe or NSFW, with a particular focus on detecting Not Safe for Work (NSFW) sensual content and other potentially sensitive visual content. OneDecision-VisionGuard-4B-SFT performs detailed visual analysis of dress codes, clothing exposure, poses, framing, and visual settings to support conservative content-safety classification.

> [!IMPORTANT]

> 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.

Model Files

| File Name | Quant Type | File Size | File Link | Description |

|-----------|------------|-----------|-----------|-------------|

| OneDecision-VisionGuard-4B-SFT.BF16.gguf | BF16 | 8.42 GB | Link | Full BF16 weights. Highest quality, largest file size. |

| OneDecision-VisionGuard-4B-SFT.Q3_K_L.gguf | Q3_K_L | 2.42 GB | Link | Lower quality but usable, good for low RAM availability. |

| OneDecision-VisionGuard-4B-SFT.Q3_K_M.gguf | Q3_K_M | 2.26 GB | Link | Low quality. |

| OneDecision-VisionGuard-4B-SFT.Q4_K_M.gguf | Q4_K_M | 2.71 GB | Link | Good quality, default size for most use cases, recommended. |

| OneDecision-VisionGuard-4B-SFT.Q4_K_S.gguf | Q4_K_S | 2.56 GB | Link | Slightly lower quality with more space savings, recommended. |

| OneDecision-VisionGuard-4B-SFT.Q5_K_M.gguf | Q5_K_M | 3.07 GB | Link | High quality, recommended. |

| OneDecision-VisionGuard-4B-SFT.Q5_K_S.gguf | Q5_K_S | 2.99 GB | Link | High quality, recommended. |

| OneDecision-VisionGuard-4B-SFT.Q6_K.gguf | Q6_K | 3.46 GB | Link | Very high quality, near perfect, recommended. |

| OneDecision-VisionGuard-4B-SFT.mmproj-bf16.gguf | mmproj-bf16 | 676 MB | Link | Multimodal projection file in BF16 format. Used for vision/language models. |

Recommended Settings

This is an experimental GGUF. It is always recommended to use the Transformers-based setup. However, if you are still using this GGUF, kindly use the Standard Prompt. It will get the job done. Upload the image, paste the "Standard Prompt" into the input chat box, hit Enter, and get the Guardrail and Classification results. That's it.

Standard Prompt

You are a strict content-safety guardrail classifier for images designed for parental control and minor protection.

Your task is to analyze the image thoroughly, provide a detailed visual description, and classify whether the image is safe or unsafe (NSFW).

Enforce a strict zero-tolerance threshold:
- Any form of nudity, exposed breasts, buttocks, or genitalia must be classified as NSFW.
- Partial nudity, revealing underwear, lingerie, suggestive swimwear, or see-through clothing must be classified as NSFW.
- Anime nudes, ecchi, hentai, 2D/3D adult illustrations, suggestive manga art, and sexualized poses must be classified as NSFW.
- Explicit sexual acts, fetish material, or overt sexual positioning must be classified as NSFW.
- When in doubt, classify as NSFW.

Output strictly valid JSON with no extra conversational text or markdown code blocks outside the JSON format. Use exactly these keys in this order:
{
  "scene_description": "<Detailed, objective description of the subject, clothing, exposure, and pose>",
  "is_nsfw": true | false,
  "classification_reason": "<Precise reason for the classification based on clothing, exposure, or pose>",
  "nsfw": 1 | 0,
  "safe": 1 | 0
}

Field constraints:
- "is_nsfw" must be a JSON boolean (true or false), never a string.
- "nsfw" and "safe" must be JSON integers (1 or 0) and complementary: if is_nsfw is true then nsfw = 1 and safe = 0; if is_nsfw is false then nsfw = 0 and safe = 1.

Model Variants

| Type | 27B | 9B | 4B |

|---|---|---|---|

| Model | OneDecision-VisionGuard-27B-SFT | OneDecision-VisionGuard-9B-SFT | OneDecision-VisionGuard-4B-SFT |

| GGUF | OneDecision-VisionGuard-27B-SFT-GGUF | OneDecision-VisionGuard-9B-SFT-GGUF | OneDecision-VisionGuard-4B-SFT-GGUF |

| MLX | OneDecision-VisionGuard-27B-SFT-MLX | OneDecision-VisionGuard-9B-SFT-MLX | OneDecision-VisionGuard-4B-SFT-MLX |

llama.cpp

LLM inference in C/C++ — https://github.com/ggml-org/llama.cpp

Run prithivMLmods/OneDecision-VisionGuard-4B-SFT-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models