DavidAU/Qwen3.5-9B-Star-Trek-TNG-DS9-Heretic-Uncensored-Thinking

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 19, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

DavidAU/Qwen3.5-9B-Star-Trek-TNG-DS9-Heretic-Uncensored-Thinking is a 9 billion parameter Qwen 3.5 dense model fine-tuned by DavidAU with a 32768 token context length. This model is specifically trained on Star Trek TNG/DS9 datasets, imbuing it with extensive 'trek' knowledge and a distinct 'trekie' persona. It features a 'Heretic' uncensored design, meaning it will not refuse requests and can generate explicit content when directed, making it suitable for creative writing and roleplay scenarios requiring specific thematic knowledge and unrestricted output.

Loading preview...

DavidAU/Qwen3.5-9B-Star-Trek-TNG-DS9-Heretic-Uncensored-Thinking

This model is a 9 billion parameter Qwen 3.5 dense model, fine-tuned by DavidAU using proprietary Star Trek: The Next Generation (TNG) and Deep Space Nine (DS9) datasets. The primary goal of this fine-tuning was to instill the model with deep 'trek' knowledge and a 'trekie' personality, influencing its thinking process, prose style, word choice, and dialogue generation. Notably, this model is a "Heretic" version, meaning it is fully uncensored and designed to fulfill user requests without refusal, including generating explicit or graphic content when explicitly directed.

Key Capabilities

  • Star Trek Themed Generation: Excels at generating content, dialogue, and narratives within the Star Trek TNG/DS9 universe, adopting a 'trekie' persona.
  • Uncensored Output: Designed to produce content without refusal, including potentially explicit, graphic, or slang-filled responses when prompted.
  • Altered Thinking & Prose: Exhibits a slightly altered thinking process, leading to more concise and "sure of itself" generations, along with distinct prose and dialogue styles.
  • Vision Capabilities: Tested and confirmed to work with new training for image inputs.
  • Improved Benchmarks: Training has resulted in raised benchmarks across various metrics compared to the base Qwen3.5-9B model, particularly in areas like arc, boolq, obkqa, and piqa.

Good For

  • Creative Writing & Roleplay: Ideal for generating fan fiction, character dialogue, or interactive roleplay scenarios within the Star Trek universe, especially those requiring uncensored or specific thematic content.
  • Thematic Content Generation: Users needing a model with specialized knowledge and a distinct voice for Star Trek TNG/DS9 lore.
  • Unrestricted Content Creation: Developers and users who require a model that will not filter or refuse to generate content based on typical safety guidelines, provided explicit direction is given.
  • Exploration of Model Behavior: For researchers interested in studying the effects of fine-tuning on model personality, style, and censorship behavior.

Finetunes
2 models