MergekitCloud/mergekit-14
MergekitCloud/mergekit-14 is an 8 billion parameter language model created by MergekitCloud, built upon the Llama 3.1 architecture. This model is a merge of several pre-trained models, including Undi95/Llama3-Unholy-8B-OAS, ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.3, and Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2, using the Model Stock merge method with vicgalle/Humanish-Roleplay-Llama-3.1-8B as its base. It is specifically optimized for roleplay and uncensored conversational applications, offering a context length of 8192 tokens.
Loading preview...
Overview
MergekitCloud/mergekit-14 is an 8 billion parameter language model derived from the Llama 3.1 family, specifically designed for enhanced conversational and roleplay capabilities. It was constructed using the advanced Model Stock merge method, integrating multiple specialized models into a cohesive unit. The base model for this merge was vicgalle/Humanish-Roleplay-Llama-3.1-8B, which was combined with Undi95/Llama3-Unholy-8B-OAS, ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.3, and Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2.
Key Characteristics
- Architecture: Llama 3.1-based, 8 billion parameters.
- Merge Method: Utilizes the Model Stock technique for combining diverse model strengths.
- Component Models: Integrates models known for roleplay and uncensored content generation.
- Context Length: Supports an 8192-token context window.
Use Cases
This model is particularly well-suited for applications requiring:
- Roleplay Scenarios: Generating dynamic and engaging character interactions.
- Creative Writing: Assisting with narrative development and dialogue generation.
- Uncensored Content Generation: Providing responses without typical content filters, as indicated by its merged components.
- Conversational AI: Developing chatbots or virtual assistants with a focus on nuanced and unrestricted dialogue.