IIEleven11/Kalypso

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kPublished:Jun 10, 2025Architecture:Transformer0.0K Warm

Kalypso is a 12 billion parameter language model developed by IIEleven11, based on the Mistral-Nemo-12B architecture. It is fine-tuned using DPO (Direct Preference Optimization) on extensive roleplay and storytelling data. This model is specifically designed for generating uncensored, immersive roleplaying and narrative content, excelling in creative writing tasks.

Loading preview...

Kalypso-12b: A Specialized Roleplay and Storytelling Model

Kalypso-12b, developed by IIEleven11, is a 12 billion parameter language model built upon the Mistral-Nemo-12B base architecture. It has undergone extensive fine-tuning using Direct Preference Optimization (DPO) on a diverse dataset of roleplay and storytelling data.

Key Capabilities

  • Specialized for Creative Narratives: Kalypso is specifically engineered to excel in generating engaging roleplay scenarios and detailed storytelling content.
  • Uncensored Output: The model is designed to be completely uncensored, offering unrestricted creative freedom for users.
  • DPO Fine-tuning: Leveraging DPO, Kalypso is optimized to produce high-quality, coherent, and contextually relevant responses for interactive narrative experiences.

Good for

  • Roleplaying: Ideal for creating dynamic and immersive roleplay interactions.
  • Storytelling: Excellent for generating creative narratives, character dialogues, and plot developments.
  • Unrestricted Content Generation: Suitable for use cases requiring uncensored and free-form text generation in creative contexts.