Qwen3VL-8B-Uncensored-HauhauCS-Balanced

Qwen3VL-8B uncensored by HauhauCS.

About

No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals.

These are meant to be the best lossless uncensored models out there.

Balanced vs Aggressive

This is the Balanced variant with moderate uncensoring. Best for agentic coding and reliability-critical tasks.

For stronger uncensoring when this variant refuses too much, use the Aggressive variant instead.

Downloads

File	Quant	Size
Qwen3VL-8B-Uncensored-HauhauCS-Balanced-BF16.gguf	BF16	16 GB
Qwen3VL-8B-Uncensored-HauhauCS-Balanced-Q8_0.gguf	Q8_0	8.2 GB
Qwen3VL-8B-Uncensored-HauhauCS-Balanced-Q6_K.gguf	Q6_K	6.3 GB
Qwen3VL-8B-Uncensored-HauhauCS-Balanced-Q4_K_M.gguf	Q4_K_M	4.7 GB
Qwen3VL-8B-Uncensored-HauhauCS-Balanced-mmproj-f16.gguf	mmproj	1.1 GB

Specs

• 8B parameters
• 256K context
• Vision-language model (requires mmproj file for image input)
• Based on Qwen3-VL-8B

Usage

Works with llama.cpp, LM Studio, koboldcpp, etc.

For vision capabilities, load both the main model and the mmproj file.

llama.cpp example:

./llama-cli -m Qwen3VL-8B-Uncensored-HauhauCS-Balanced-Q4_K_M.gguf \
  --mmproj Qwen3VL-8B-Uncensored-HauhauCS-Balanced-mmproj-f16.gguf \
  --image your_image.jpg \
  -p "Describe this image"