| Property | Value |
|---|---|
| Model Size | 12B parameters |
| Format | GGUF |
| Author | Delta-Vector |
| Base Models | Rei-12B, Francois-Huali-12B |
| Merge Method | SLERP (t=0.2) |
What is Archaeo-12B-GGUF?
Archaeo-12B-GGUF is a specialized language model created through a strategic merge of Rei-12B and Francois-Huali-12B models using SLERP interpolation. The model is specifically optimized for roleplay and creative writing applications, featuring ChatML formatting support and static GGUF quantization for efficient deployment.
Implementation Details
The model utilizes a SLERP merge method with a t-value of 0.2, balanced between the base models. It's implemented with bfloat16 dtype and maintains the tokenizer from the Rei-12B base model. The architecture supports both GGUF and EXL2 quantization formats for flexible deployment options.
- ChatML formatting support for structured conversations
- SLERP interpolation with 0.2 weighting
- Multiple quantization options (GGUF and EXL2)
- bfloat16 precision implementation
Core Capabilities
- Specialized in roleplay and creative writing tasks
- Structured conversation handling through ChatML
- Optimized for efficient deployment with GGUF quantization
- Enhanced creative text generation capabilities
Frequently Asked Questions
Q: What makes this model unique?
The model's unique strength lies in its specialized merge of Rei-12B and Francois-Huali-12B, specifically optimized for creative writing and roleplay scenarios. The SLERP merge method with a 0.2 t-value creates a balanced combination of both models' capabilities.
Q: What are the recommended use cases?
Archaeo-12B-GGUF is primarily designed for creative writing and roleplaying applications. It excels in generating narrative content, character interactions, and structured creative conversations using ChatML formatting.