Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
Paper
• 2203.05482 • Published
• 7
This is a merge of pre-trained language models created using mergekit.
This model was merged using the Linear merge method.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: Tibogoss/Qwen3-8B-LoRA_v3-drug
parameters:
weight: 0.333333
- model: Tibogoss/Qwen3-8B-LoRA_v3-ge
parameters:
weight: 0.333333
- model: Tibogoss/Qwen3-8B-LoRA_v2
parameters:
weight: 0.333333
merge_method: linear
dtype: bfloat16