SvalTek/MQ-Coldbrew-Base-Test1

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 4, 2026Architecture:Transformer Featherless Exclusive Cold

SvalTek/MQ-Coldbrew-Base-Test1 is a 7.6 billion parameter language model created by SvalTek, merged using the TIES method with SvalTek/Qwen2.5-ColdBrew as its base. This model integrates components from bunnycore/Qwen-2.5-7b-TitanFusion-v3, focusing on combining their respective strengths. It is designed for general language generation tasks, leveraging the Qwen2.5 architecture for robust performance.

Loading preview...

Overview

SvalTek/MQ-Coldbrew-Base-Test1 is a 7.6 billion parameter language model developed by SvalTek. It was created using the TIES merge method from MergeKit, combining the strengths of existing pre-trained models.

Merge Details

This model uses SvalTek/Qwen2.5-ColdBrew as its foundational base model. It integrates parameters from bunnycore/Qwen-2.5-7b-TitanFusion-v3 and SvalTek/Qwen2.5-ColdBrew, with specific weighting and density configurations to optimize their combined performance. The merge process was configured to normalize parameters and was performed using bfloat16 data type.

Key Characteristics

  • Architecture: Based on the Qwen2.5 family of models.
  • Parameter Count: 7.6 billion parameters.
  • Merge Method: Utilizes the TIES (Trimmed, Iterative, and Selective) merging technique.
  • Base Model: Built upon SvalTek/Qwen2.5-ColdBrew.

Intended Use

This model is suitable for general-purpose language generation and understanding tasks, benefiting from the combined capabilities of its constituent models. Developers can leverage its merged architecture for applications requiring a robust 7.6B parameter model.