VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored-GGUF
VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored is a multimodal safety classifier built on top of Qwen/Qwen3.5-4B. The model was trained on a mixture of approximately 10,000 video safety and scene-reasoning samples to analyze video content and classify potentially unsafe content across predefined safety categories. The model is designed to generate a structured DESCRIPTION, EXPLANATION, and GUARDRAIL output, making it suitable for video content filtering, safety evaluation, and multimodal guardrail research.
[!NOTE] This model is an experimental release and may generate unexpected classifications or reasoning artifacts in certain scenarios. Safety classifications should be treated as model predictions rather than definitive judgments.
Model Files
| File Name | Quant Type | File Size | File Link |
|---|---|---|---|
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.BF16.gguf | BF16 | 8.42 GB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.F16.gguf | F16 | 8.42 GB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.Q4_K_M.gguf | Q4_K_M | 2.71 GB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.Q5_K_M.gguf | Q5_K_M | 3.07 GB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.Q8_0.gguf | Q8_0 | 4.48 GB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.mmproj-bf16.gguf | mmproj-bf16 | 676 MB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.mmproj-f16.gguf | mmproj-f16 | 676 MB | Download |
| VideoGuard-Qwen3.5-4B-Safety-RL-Uncensored.mmproj-q8_0.gguf | mmproj-q8_0 | 367 MB | Download |
llama.cpp
LLM inference in C/C++ — https://github.com/ggml-org/llama.cpp