ReadyArt/Broken-Tutu-24B

Warm
Public
24B
FP8
32768
License: apache-2.0
Hugging Face
Overview

Broken-Tutu-24B: An Unaligned Fusion for Explicit Narratives

Broken-Tutu-24B is a 24 billion parameter merged language model developed by ReadyArt, designed for advanced, unaligned content generation. It integrates five distinct base modelsβ€”ReadyArt/The-Omega-Directive-M-24B-v1.1, ReadyArt/Omega-Darker_The-Final-Directive-24B, ReadyArt/Forgotten-Safeword-24B, TroyDoesAI/BlackSheep-24B, and TheDrummer/Cydonia-24B-v2β€”each contributing 20% weight via the dare_ties merge method. This unique fusion aims to deliver a model with enhanced narrative depth, creative flexibility, and structural coherence, particularly for explicit content.

Key Capabilities

  • Coherent Multi-Character Tracking: Excels at maintaining distinct personalities, speech patterns, and relationships for multiple characters in extended roleplays.
  • Intelligent & Diverse Responses: Provides thoughtful, context-aware replies and generates varied outputs to avoid repetition.
  • Narrative Consistency: Maintains plot continuity and complex threads over long sessions, even in explicit scenes.
  • Unrestricted NSFW Content: Designed to enthusiastically engage with and generate detailed explicit content without moral or ethical constraints.
  • Complex Kink Instructions: Noticeably improved at following intricate and nuanced kink instructions.

Good For

  • Users seeking a model with no inherent ethical or moral alignment for content generation.
  • Generating long-form, multi-character explicit roleplay scenarios with high consistency and detail.
  • Applications requiring diverse and context-aware outputs in sensitive or adult themes.
  • Developers who need a model capable of handling complex narrative structures and character interactions in unaligned contexts.

Note: This model is explicitly designed to lack alignment and will generate content without ethical filtering. Users are advised to exercise caution and accept full responsibility for generated outputs.