It seems better than DataSoul/QAQ-32B-merge2. But it is still not a 'thinking' model.

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterated as a base.

Models Merged

The following models were included in the merge:

DataSoul/QAQ-32B-merge2

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: DataSoul/QAQ-32B-merge2
    parameters:
      weight: 0.382
      density: 0.618
  - model: zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterated
    parameters:
      weight: 1
      density: 1
merge_method: ties
base_model: zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterated
tokenizer_source: zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterated
parameters:
  normalize: true
  int8_mask: true
dtype: float16