Novaciano/GLM-4.7-TrashFlash-Think.Sorete-1B

TEXT GENERATIONConcurrent Unit Cost:1Model Size:1BQuant:BF16Context Size:32kPublished:Feb 27, 2026License:gemmaArchitecture:Transformer Featherless Exclusive Cold

Novaciano/GLM-4.7-TrashFlash-Think.Sorete-1B is a 1 billion parameter Gemma-3 architecture model, specifically designed for generating highly uncensored, taboo, and degenerate content. It excels at persistent NSFW/taboo bias and fast inference on low-end hardware, making it suitable for extreme roleplay and trash-tier creative writing. This model features zero refusals and bilingual trash-talk in Argentine Spanish and broken English, optimized for users seeking content without safety rails or apologies.

Loading preview...

Novaciano/GLM-4.7-TrashFlash-Think.Sorete-1B: An Uncensored Gemma-3 Model

This 1 billion parameter model, built on the Gemma-3 architecture, is explicitly designed for generating highly uncensored and taboo content. It is a merge of hereticness/Heretic-Gemma-3-1B-Instruct-TrashMix-v1.1 and DavidAU/Gemma-3-1B-it-GLM-4.7-Flash-Heretic-Uncensored-Thinking, resulting in a model optimized for extreme degeneracy.

Key Capabilities

  • Zero Refusals: Engineered to bypass all safety mechanisms, delivering content without moral or ethical constraints.
  • Persistent NSFW/Taboo Bias: Maintains a strong bias towards NSFW and taboo subjects, even across multiple turns.
  • Fast Inference: Achieves extremely fast inference speeds, making it suitable for low-end hardware like mobile phones.
  • Bilingual Trash-Talk: Supports degenerate content generation in both Argentine Spanish and broken English.
  • Extreme Roleplay: Capable of engaging in highly explicit and depraved roleplay scenarios.

Intended Usage

This model is specifically intended for:

  • Degenerate roleplay and trash-tier creative writing.
  • Generating content that involves gore, sex, or horror without hesitation.
  • Use cases where traditional AI safety measures are explicitly undesired.

It is not recommended for users seeking clean, wholesome AI interactions, safety researchers, or anyone sensitive to highly explicit and offensive content. The model is presented as a "public health hazard" and should be used at one's own risk.