merged_qwen_omni_dare_ties_3B

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using Qwen/Qwen2.5-Omni-3B as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

# mergekit ์„ค์ • ํŒŒ์ผ: Qwen/Qwen3-8B ์™€ Qwen/Qwen2.5-Omni-7B ๋ฅผ DARE_TIES ๋ฐฉ์‹์œผ๋กœ ๋ณ‘ํ•ฉ

# ํ•ฉ์น  ๋ชจ๋ธ ๋ชฉ๋ก
models:
  - model: Qwen/Qwen2-Audio-7B
    parameters:
      weight: 1.0
      density: 1.0

  - model: Qwen/Qwen2.5-Omni-3B
    parameters:
      weight: 1.0
      density: 1.0

merge_method: dare_ties
base_model: Qwen/Qwen2.5-Omni-3B # ๋˜๋Š” Qwen/Qwen3-8B

parameters:
  density: 0.5 # ์ด ๊ฐ’์„ ์กฐ์ •ํ•˜๋ฉฐ ์‹คํ—˜ํ•˜์„ธ์š” (์˜ˆ: 0.2, 0.7 ๋“ฑ)
  normalize: false
  int8_mask: true

dtype: bfloat16
Downloads last month
17
Safetensors
Model size
635M params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for ldhldh/merged-qwen-omni-dare-3B

Merge model
this model