DarkArtsForge/Paimon-24B-v1
Paimon-24B-v1 by DarkArtsForge is a 24 billion parameter merged language model built upon the MistralForCausalLM architecture, using mistralai/Mistral-Small-24B-Instruct-2501 as its base. This model is specifically engineered through the DELLA merge method to produce narratives and roleplay content that can include violent and graphic erotic themes. It is optimized for creative and unfiltered text generation, requiring careful system prompt adjustments for appropriate use.
Loading preview...
Paimon-24B-v1: A Merged Model for Unfiltered Narrative Generation
Paimon-24B-v1 is a 24 billion parameter language model developed by DarkArtsForge, utilizing the MistralForCausalLM architecture. It is a merge of several pre-trained models, with mistralai/Mistral-Small-24B-Instruct-2501 serving as its base. The model was created using the DELLA merge method, integrating components from various models like c4tdr0ut/grok-oss-Apollyon-24B, Casual-Autopsy/The-True-Abomination-24B, and DavidAU/Dolphin-Mistral-GLM-4.7-Flash-24B-Venice-Edition-Thinking-Uncensored.
Key Capabilities
- Specialized Content Generation: Designed to produce narratives and roleplay content that may include violent and graphic erotic themes.
- Merge Method: Employs the DELLA merge method for combining diverse model strengths.
- Iterative Development: The model has undergone multiple iterations (v1, v2, v3) with adjustments to the merged components, excluding models that distorted the merge or reactivated positivity bias in later versions.
Good For
- Creative Writing: Ideal for generating highly specific and potentially explicit narrative content.
- Roleplay Scenarios: Suited for roleplaying applications where unfiltered and graphic themes are desired.
Note: Users are advised to adjust their system prompts accordingly and use the Mistral Tekken template for optimal results, given the model's capacity for generating sensitive content.