README.md

---
license: apache-2.0
language:
- en
base_model: Suu/Klear-Reasoner-8B
datasets:
- Suu/KlearReasoner-MathSub-30K
- Suu/KlearReasoner-CodeSub-15K
metrics:
- accuracy
tags:
- mlx
---

# hobaratio/Klear-Reasoner-8B-mlx-8Bit

The Model [hobaratio/Klear-Reasoner-8B-mlx-8Bit](https://huggingface.co/hobaratio/Klear-Reasoner-8B-mlx-8Bit) was converted to MLX format from [Suu/Klear-Reasoner-8B](https://huggingface.co/Suu/Klear-Reasoner-8B) using mlx-lm version **0.26.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("hobaratio/Klear-Reasoner-8B-mlx-8Bit")

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)
```