README.md

---
base_model:
- Bingguang/FunReason-MT
- openfree/Darwin-Qwen3-4B
- ricdomolm/mini-coder-4b
- minchyeom/Qwaifu
tags:
- merge
- mergekit
- lazymergekit
- Bingguang/FunReason-MT
- openfree/Darwin-Qwen3-4B
- ricdomolm/mini-coder-4b
- minchyeom/Qwaifu
---

# Qwen3-4B-Mini-Merge

Qwen3-4B-Mini-Merge is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Bingguang/FunReason-MT](https://huggingface.co/Bingguang/FunReason-MT)
* [openfree/Darwin-Qwen3-4B](https://huggingface.co/openfree/Darwin-Qwen3-4B)
* [ricdomolm/mini-coder-4b](https://huggingface.co/ricdomolm/mini-coder-4b)
* [minchyeom/Qwaifu](https://huggingface.co/minchyeom/Qwaifu)

## 🧩 Configuration

```yaml

models:
  - model: Bingguang/FunReason-MT
    parameters:
      density: 0.3
      weight: 0.3
  - model: openfree/Darwin-Qwen3-4B
    parameters:
      density: 0.3
      weight: 0.3
  - model: ricdomolm/mini-coder-4b
    parameters:
      density: 0.3
      weight: 0.3
  - model: minchyeom/Qwaifu
    parameters:
      density: 0.5
      weight: 0.5
merge_method: ties
base_model: huihui-ai/Huihui-Qwen3-4B-Instruct-2507-abliterated
parameters:
  normalize: true
dtype: float16

```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "bunnycore/Qwen3-4B-Mini-Merge"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```