---
library_name: transformers
tags:
- ablated
- ablation
- abliterated
- uncensored
- nani
- reasoning
- qwen
- deepseek
- mlx
- mlx-my-repo
license: mit
base_model: NaniDAO/deepseek-r1-qwen-2.5-32B-ablated
---

# hentograf/deepseek-r1-qwen-2.5-32B-ablated-Q4-mlx

The Model [hentograf/deepseek-r1-qwen-2.5-32B-ablated-Q4-mlx](https://huggingface.co/hentograf/deepseek-r1-qwen-2.5-32B-ablated-Q4-mlx) was converted to MLX format from [NaniDAO/deepseek-r1-qwen-2.5-32B-ablated](https://huggingface.co/NaniDAO/deepseek-r1-qwen-2.5-32B-ablated) using mlx-lm version **0.20.5**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("hentograf/deepseek-r1-qwen-2.5-32B-ablated-Q4-mlx")

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)
```