Sorihon/Magistry-Painted-Cydonia-24B
Sorihon/Magistry-Painted-Cydonia-24B is a 24 billion parameter language model created by Sorihon, formed by merging three distinct 24B models using the Karcher Mean method. This model combines the strengths of sophosympatheia/Magistry-24B-v1.1, zerofata/MS3.2-PaintedFantasy-v4.1-24B, and TheDrummer/Cydonia-24B-v4.2.0. It is designed to leverage the collective capabilities of its constituent models, offering a versatile foundation for various natural language processing tasks.
Loading preview...
Model Overview
Sorihon/Magistry-Painted-Cydonia-24B is a 24 billion parameter language model developed by Sorihon. This model is a product of a sophisticated merge operation, combining the strengths of multiple pre-trained language models to create a more robust and versatile foundation.
Merge Details
This model was created using the Karcher Mean merge method, a technique known for effectively blending the characteristics of different models. The merge process involved three distinct 24B parameter models:
- sophosympatheia/Magistry-24B-v1.1
- zerofata/MS3.2-PaintedFantasy-v4.1-24B
- TheDrummer/Cydonia-24B-v4.2.0
The configuration utilized a karcher merge method, with union tokenizer sourcing and auto chat template detection, outputting in bfloat16 data type. This approach aims to synthesize the unique capabilities and knowledge bases of each constituent model into a single, more capable entity.
Potential Use Cases
Given its merged nature, Magistry-Painted-Cydonia-24B is likely suitable for a broad range of applications where the individual strengths of its merged components are beneficial. While specific optimizations are not detailed, models created via Karcher Mean merging often exhibit improved generalization and performance across diverse tasks compared to their individual constituents.