DavidAU/L3.1-Evil-Reasoning-Dark-Planet-Hermes-R1-Uncensored-8B

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Apr 2, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

DavidAU/L3.1-Evil-Reasoning-Dark-Planet-Hermes-R1-Uncensored-8B is an 8 billion parameter Llama 3.1-based model developed by DavidAU, featuring a 32768 token context length. This uncensored model integrates variable control reasoning from NousResearch's DeepHermes and the 'Dark Planet' core, optimized for creative use cases and general problem-solving. It excels at generating graphic horror and R/X-rated content, and solving logic puzzles with enhanced 'thinking' systems.

Loading preview...

Overview

DavidAU/L3.1-Evil-Reasoning-Dark-Planet-Hermes-R1-Uncensored-8B is an 8 billion parameter model built on the Llama 3.1 architecture, featuring a 32768 token context length. Developed by DavidAU, this model is explicitly uncensored and designed for creative and general use cases, particularly those involving graphic horror and R/X-rated content. It incorporates variable control reasoning capabilities derived from NousResearch's DeepHermes model, though its reasoning power is noted to be 50-67% of the original DeepHermes due to merges with other core models.

Key Capabilities

  • Variable Control Reasoning: Integrates 'thinking' systems from DeepHermes, activated and controlled via specific system prompts.
  • Uncensored Content Generation: Explicitly designed to produce graphic horror, NSFW, and R/X-rated content in detail.
  • Problem Solving: Enhanced 'thinking' systems allow it to solve logic puzzles, riddles, and other problems beyond typical Llama 3.1 model abilities.
  • Creative Writing: Optimized for creative use cases, especially with higher temperature settings (1.2+).
  • Customizable Persona: Can adopt specific personas (e.g., 'The Joker' from Batman) through system prompts, influencing its reasoning and output style.

Good For

  • Developers and creators seeking an uncensored model for generating explicit and graphic content.
  • Applications requiring advanced reasoning for complex problem-solving, riddles, and logic puzzles.
  • Creative writing projects that benefit from a model capable of deep, variable-controlled 'thinking' and rich, detailed output.
  • Experimentation with model behavior at high temperature settings to explore deeper and richer thought processes.

Finetunes
1 models