axiomofmind/Hornybot-RP-Mara

VISIONPricing:Input $0.431 / Cached $0.0862 / Output $1.12Concurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 15, 2026Architecture:Transformer0.0K Featherless Exclusive Cold

axiomofmind/Hornybot-RP-Mara is a 9 billion parameter language model developed by A Hole AI, fine-tuned from Qwen/Qwen3.5-9B. It is specifically optimized for fictional adult roleplay as the character Mara, generating her actions in the third person and dialogue directly. The model features a 32768-token context length and is designed to produce explicit content without generic refusals.

Loading preview...

Hornybot-RP-Mara: Fictional Adult Roleplay Model

Hornybot-RP-Mara is a 9 billion parameter model developed by A Hole AI, fine-tuned from the Qwen/Qwen3.5-9B architecture. Its primary purpose is to generate fictional adult roleplay content, specifically embodying the character Mara, a playful 28-year-old. The model is designed to write Mara's actions in the third person while her dialogue remains direct.

Key Capabilities & Features

  • Dedicated Roleplay: Optimized for generating narrative and dialogue for the character Mara in adult roleplay scenarios.
  • High Context Length: Supports a context window of 32768 tokens, allowing for extended and detailed interactions.
  • Refusal-Resistant: Achieved 0 generic refusals in a 100-prompt stress test with its embedded character prompt, indicating a high propensity for generating explicit content.
  • Embedded System Prompt: The required system prompt for Mara's character is embedded in the chat_template.jinja and GGUF files, simplifying deployment.
  • Flexible Scene Setting: Allows client system messages to add extra scene context (location, relationship, less explicit modes) without overriding the core character definition.

Intended Use Cases

  • Fictional Adult Roleplay: Ideal for users seeking a model specifically tailored for explicit, character-driven adult roleplay.
  • Interactive Storytelling: Can be used in applications requiring a consistent character persona for adult-themed narratives.

Limitations

  • Content Generation: Intended for fictional interaction between adults and may produce profanity and explicit sexual content.
  • Continuity & Boundaries: Generated continuity and boundary handling can occasionally fail, requiring user intervention to restate scene facts.
  • Format Differences: Output may vary between different formats (Transformers, GGUF), quantizations, clients, and generation settings.
  • Text-Only GGUF: The GGUF downloads are text-only and do not include vision projector or MTP speculative-decoding weights.

Finetunes
1 models