SvalTek/MQ-Coldbrew-Base-Test0

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

SvalTek/MQ-Coldbrew-Base-Test0 is a 7.6 billion parameter language model created by SvalTek, merged using the TIES method. It is based on SvalTek/Qwen2.5-ColdBrew and incorporates Orion-zhen/Meissa-Qwen2.5-7B-Instruct, offering a 32768 token context length. This model is designed as a foundational base, combining characteristics from its merged components for general language understanding and generation tasks.

Loading preview...

Model Overview

SvalTek/MQ-Coldbrew-Base-Test0 is a 7.6 billion parameter language model developed by SvalTek. This model was created using the TIES merge method from mergekit, combining the strengths of multiple pre-trained models.

Key Capabilities

  • Merged Architecture: Built upon SvalTek/Qwen2.5-ColdBrew as its base, integrating Orion-zhen/Meissa-Qwen2.5-7B-Instruct to enhance its capabilities.
  • TIES Merge Method: Utilizes the TIES (Trimmed, Iterative, and Selective) merging technique, which is designed to combine models effectively by selectively merging parameters.
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for processing and generating longer sequences of text.
  • Bfloat16 Precision: The model is configured to use bfloat16 data type, balancing performance and memory efficiency.

Good For

  • General Language Tasks: Suitable for a wide range of natural language processing applications due to its merged foundation.
  • Further Fine-tuning: Can serve as a robust base model for subsequent fine-tuning on specific downstream tasks or datasets.
  • Exploration of Merged Models: Provides an example of a model created through advanced merging techniques, useful for researchers and developers interested in model combination strategies.

Finetunes
1 models