ReadyArt's Broken-Tutu-24B-Transgression-v2.0 is a 24 billion parameter language model fine-tuned from mistralai/Mistral-Small-24B-Instruct-2501, designed for enhanced narrative coherence and character integrity in roleplay and storytelling scenarios. It features an expanded 43 million token multi-turn conversational dataset and anti-impersonation guards, focusing on reducing explicit content while maintaining character authenticity. This model excels at long-form, multi-character scenarios and superior instruction following with complex prompts.
No reviews yet. Be the first to review!