jaspionjader/Kosmos-EVAA-Franken-stock-v42-8B

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Jan 17, 2025Architecture:Transformer0.0K Cold

The jaspionjader/Kosmos-EVAA-Franken-stock-v42-8B is an 8 billion parameter language model created by jaspionjader, utilizing the Model Stock merge method. This model was built upon jaspionjader/kstc-5-8b as a base, integrating components from jaspionjader/kstc-4-8b and jaspionjader/bbb-6. It is designed for general language tasks, leveraging its merged architecture to combine the strengths of its constituent models.

Loading preview...

Model Overview

The jaspionjader/Kosmos-EVAA-Franken-stock-v42-8B is an 8 billion parameter language model developed by jaspionjader. This model is a product of the Model Stock merge method, a technique designed to combine the capabilities of multiple pre-trained language models.

Merge Details

The model was constructed using mergekit and is based on jaspionjader/kstc-5-8b. It incorporates contributions from two additional models:

  • jaspionjader/kstc-4-8b
  • jaspionjader/bbb-6

This merging strategy aims to leverage the distinct characteristics and knowledge bases of its constituent models, potentially enhancing its overall performance and versatility for various language understanding and generation tasks. The merge was performed with bfloat16 precision.

Intended Use

Given its merged architecture, this model is suitable for a broad range of applications where a robust 8B parameter model is beneficial. Developers can explore its capabilities for tasks such as text generation, summarization, question answering, and more, benefiting from the combined strengths of its merged components.