DavidAU/Qwen3.5-9B-The-Deckard-Pro-Writer-Uncensored-Heretic
DavidAU/Qwen3.5-9B-The-Deckard-Pro-Writer-Uncensored-Heretic is a 9 billion parameter Qwen3.5-based causal language model developed by DavidAU. This model is specifically fine-tuned to eliminate 'Qwen Prose' and refusal behaviors, offering fully uncensored output. It excels in creative writing, roleplay, and generating graphic content without stylistic 'slop', while also supporting vision inputs and maintaining a 262,144 token context length.
Loading preview...
Model Overview
DavidAU/Qwen3.5-9B-The-Deckard-Pro-Writer-Uncensored-Heretic is a specialized 9 billion parameter Qwen3.5 model, developed by DavidAU, focusing on unfiltered and direct content generation. It has undergone "Deckard STRICT training" to eliminate common stylistic issues found in base Qwen models, often referred to as "Qwen Prose," and to remove refusal behaviors, making it a "HERETIC" model that generates content "no questions asked."
Key Capabilities
- Uncensored Content Generation: Designed to produce graphic, explicit, or sensitive content without refusal, requiring only minimal direction for desired intensity.
- Elimination of "Qwen Prose": Retrained to remove stylistic 'slop' such as repetitive phrases and awkward sentence structures, improving narrative quality.
- Multimodal Support: Retains the base Qwen3.5's vision capabilities, with image input tested and working, and video input supported (though not explicitly tested in this specific fine-tune).
- Extended Context Length: Supports a native context length of 262,144 tokens, extensible up to 1,010,000 tokens using YaRN scaling techniques.
What Makes This Different?
This model's primary differentiator is its explicit focus on uncensored output and the elimination of stylistic 'slop' from the base Qwen model. While other models might be uncensored, this version specifically targets and corrects the "Qwen Prose" issue, aiming for cleaner, more direct, and vivid writing. It achieves a significantly lower refusal rate (6/100) compared to the original Qwen3.5-9B (100/100).
Should I Use This?
- Use this if: You require a model for creative writing, roleplay, or generating content that is explicitly uncensored and direct, without the stylistic quirks or refusal behaviors often found in other LLMs. It's particularly suited for applications where raw, unfiltered, and vivid descriptions are paramount.
- Avoid this if: Your application requires strict content moderation, adherence to ethical guidelines, or if you prefer models that default to more conservative or filtered outputs.