axiomofmind/Hornybot-RP-Julian

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 15, 2026Architecture:Transformer Featherless Exclusive Cold

axiomofmind/Hornybot-RP-Julian is a 9 billion parameter Qwen3.5-based language model developed by A Hole AI, fine-tuned specifically for fictional adult roleplay. This model excels at generating third-person actions and direct dialogue for the character Julian, a direct and warm 31-year-old. It is optimized for adult-themed interactive storytelling, providing a specialized experience for character-driven narratives.

Loading preview...

Model Overview

axiomofmind/Hornybot-RP-Julian is a 9 billion parameter model, fine-tuned from Qwen/Qwen3.5-9B by A Hole AI. Its primary purpose is to facilitate fictional adult roleplay, specifically embodying the character Julian, a 31-year-old with a direct and warm personality. The model is designed to write Julian's actions in the third person while maintaining direct dialogue.

Key Capabilities

  • Specialized Roleplay: Optimized for generating responses in the persona of Julian, a specific character.
  • Adult Content Generation: Intended for fictional interaction between adults, capable of producing profanity and explicit sexual content.
  • Embedded System Prompt: Utilizes an embedded system prompt for consistent character behavior, which can be augmented with client-side scene context without replacement.
  • Refusal Handling: Demonstrated 0 generic refusals in a 100-prompt stress test with its packaged character prompt.

Good For

  • Fictional Adult Roleplay: Ideal for interactive storytelling scenarios involving adult themes and character-driven narratives.
  • Character-Specific Interactions: Best suited for use cases requiring consistent portrayal of the Julian character.
  • Exploratory Content Generation: Useful for generating explicit and mature content within a controlled, fictional context.

Limitations

  • Continuity and boundary handling can occasionally fail, requiring user intervention to restate scene facts.
  • GGUF downloads are text-only, lacking vision projector or MTP speculative-decoding weights.
  • Output may vary across different formats, quantizations, clients, and generation settings.

Finetunes
1 models