DavidAU/Llama3.3-8B-Instruct-Thinking-Heretic-Uncensored-Claude-4.5-Opus-High-Reasoning is an 8 billion parameter Llama 3.3-based instruction-tuned model, adjusted for a 128K context window. It has been de-censored ("Heretic'ed") and further trained with a Claude 4.5-Opus High Reasoning dataset, creating a hybrid instruct/thinking model. This model excels at generating detailed, high-reasoning responses and uncensored content, making it suitable for complex narrative generation and creative tasks.
No reviews yet. Be the first to review!