sequelbox/Qwen3-14B-Esper3Mix

This is a merge of pre-trained language models created using mergekit, combining several Qwen 3 14b finetunes to maximize reasoning performance.

Merge Details

Merge Method

This model was merged using the DELLA merge method using Qwen/Qwen3-14B as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

merge_method: della
dtype: bfloat16
parameters:
  normalize: true
models:
  - model: ValiantLabs/Qwen3-14B-Esper3
    parameters:
      density: 0.25
      weight: 0.4
  - model: ValiantLabs/Qwen3-14B-Cobalt2
    parameters:
      density: 0.25
      weight: 0.25
  - model: DMindAI/DMind-1-mini
    parameters:
      density: 0.25
      weight: 0.25
  - model: soob3123/GrayLine-Qwen3-14B
    parameters:
      density: 0.25
      weight: 0.25
base_model: Qwen/Qwen3-14B