---
base_model: AlejandroOlmedo/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx
datasets: DigitalLearningGmbH/MATH-lighteval
library_name: transformers
model_name: DeepSeek-R1-Distill-Qwen-7B-GRPO_Math
tags:
- generated_from_trainer
- open-r1
- trl
- grpo
- mlx
- mlx-my-repo
- mlx
- mlx-my-repo
licence: license
---

# introvoyz041/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx-mlx-4Bit

The Model [introvoyz041/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx-mlx-4Bit](https://huggingface.co/introvoyz041/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx-mlx-4Bit) was converted to MLX format from [AlejandroOlmedo/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx](https://huggingface.co/AlejandroOlmedo/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx) using mlx-lm version **0.28.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("introvoyz041/DeepSeek-R1-Distill-Qwen-7B-GRPO_Math-8bit-mlx-mlx-4Bit")

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)
```