---
base_model: JosephNguyen/Qwen2.5-7B-Instruct-reasoning-finetuned
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - qwen2
  - trl
  - sft
  - mlx
  - mlx-my-repo
license: mit
language:
  - vi
---

# itzmealvin/MLX_Qwen2.5-7B-IT-Q8-FT-RSN

The Model [itzmealvin/MLX_Qwen2.5-7B-IT-Q8-FT-RSN](https://huggingface.co/itzmealvin/MLX_Qwen2.5-7B-IT-Q8-FT-RSN) was converted to MLX format from [JosephNguyen/Qwen2.5-7B-Instruct-reasoning-finetuned](https://huggingface.co/JosephNguyen/Qwen2.5-7B-Instruct-reasoning-finetuned) using mlx-lm version **0.22.3**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("itzmealvin/MLX_Qwen2.5-7B-IT-Q8-FT-RSN")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```