Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Paper
•
2311.03099
•
Published
•
30
This is a merge of pre-trained language models created using mergekit.
This model was merged using the DARE TIES merge method using Squelching-Fantasies-qw3-14B as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
merge_method: dare_ties
base_model: Squelching-Fantasies-qw3-14B
models:
- model: Squelching-Fantasies-qw3-14B
parameters:
weight: 0.4
- model: Vitus-Qwen3-14B
parameters:
weight: 0.35
- model: The-Omega-Directive-Qwen3-14B-v1.1
parameters:
weight: 0.25
parameters:
density: 0.4
Totally Free + Zero Barriers + No Login Required