Achieve the Optimal Merged Model by Using One Basic Model and Two Fine-tuned Models!

Name: Qwen2.5-14B-YOYO-super
Availability: InStock

What is the best way to merge one base model and two fine-tuned models?

This might be the best answer at the present stage!

Qwen2.5-7B-YOYO-super

Qwen2.5-14B-YOYO-super

This is not a whim release, but the optimal result of countless merging experiments!

Here is the formula for the previous generation:

yaml

models:  
  - model: Qwen/Qwen2.5-14B-Instruct  
    parameters:  
      density: 1  
      weight: 1  
      lambda: 0.9
  - model: Qwen/Qwen2.5-14B-Instruct-1M  
    parameters:  
      density: 1  
      weight: 1  
      lambda: 0.9  
merge_method: della  
base_model: Qwen/Qwen2.5-14B  
parameters:  
  density: 1  
  weight: 1  
  lambda: 0.9  
  normalize: true  
  int8_mask: true  
dtype: bfloat16  
tokenizer_source: base

It was widely used in the merging process of the previous generation of models.

However, there are some deficiencies:

1.There is relatively little retention of knowledge of the basic model.

2.The mathematical and coding abilities have declined.

And here is the formula for this generation:

yaml

models:
  - model: Qwen/Qwen2.5-14B-instruct
    parameters:
      density: 1 
      weight: 1
      lambda: 0.9
merge_method: della
base_model: Qwen/Qwen2.5-14B
parameters:
  density: 1
  weight: 1
  lambda: 0.9
  normalize: true
  int8_mask: true
dtype: float16
tokenizer_source: base
name: Qwen2.5-14B-della

yaml

models:
  - model: Qwen/Qwen2.5-14B-instruct-1M
    parameters:
      density: 1 
      weight: 1
      lambda: 0.9
merge_method: della
base_model: Qwen/Qwen2.5-14B
parameters:
  density: 1
  weight: 1
  lambda: 0.9
  normalize: true
  int8_mask: true
dtype: float16
tokenizer_source: base
name: Qwen2.5-14B-della-1M

yaml

models:
  - model: Qwen/Qwen2.5-14B-instruct
    parameters:
      density: 1 
      weight: 1
merge_method: ties
base_model: Qwen/Qwen2.5-14B
parameters:
  density: 1
  weight: 1
  normalize: true
  int8_mask: true
dtype: float16
tokenizer_source: base
name: Qwen2.5-14B-ties

yaml

models:
  - model: Qwen/Qwen2.5-14B-instruct-1M
    parameters:
      density: 1 
      weight: 1
merge_method: ties
base_model: Qwen/Qwen2.5-14B
parameters:
  density: 1
  weight: 1
  normalize: true
  int8_mask: true
dtype: float16
tokenizer_source: base
name: Qwen2.5-14B-ties-1M

yaml

merge_method: model_stock
base_model: Qwen/Qwen2.5-14B
models:
  - model: mergekit-community/Qwen2.5-14B-della
  - model: mergekit-community/Qwen2.5-14B-della-1M
  - model: mergekit-community/Qwen2.5-14B-ties
  - model: mergekit-community/Qwen2.5-14B-ties-1M
  - model: Qwen/Qwen2.5-14B-instruct-1M
  - model: Qwen/Qwen2.5-14B-instruct
tokenizer_source: base
int8_mask: true
normalize: true
dtype: float16

Except for a slight decrease in instruction following, significant improvements have been achieved in all other aspects.

This formula will also be used in the development of the next generation of YOYO models.

YOYO-AI not only releases merged models with excellent performance but also publishes a complete and high-quality model merging formula, hoping to promote the progress of model merging technology in the open-source community with this!

YOYO-AI/
Qwen2.5-14B-YOYO-super

Achieve the Optimal Merged Model by Using One Basic Model and Two Fine-tuned Models!

This might be the best answer at the present stage!

And here is the formula for this generation:

If you can use this formula when merging models, it will be the greatest support for YOYO-AI!

YOYO-AI/Qwen2.5-14B-YOYO-super

Achieve the Optimal Merged Model by Using One Basic Model and Two Fine-tuned Models!

This might be the best answer at the present stage!

And here is the formula for this generation:

If you can use this formula when merging models, it will be the greatest support for YOYO-AI!

YOYO-AI/
Qwen2.5-14B-YOYO-super