---
license: apache-2.0
language:
- en
- zh
library_name: transformers
pipeline_tag: text-generation
tags:
- llm
- nanbeige
- heretic
- uncensored
- decensored
- abliterated
- mlx
base_model: heretic-org/Nanbeige4.1-3B-heretic
---

# usermma/Nanbeige4.1-3B-heretic-mlx-5bit

The Model [usermma/Nanbeige4.1-3B-heretic-mlx-5bit](https://huggingface.co/usermma/Nanbeige4.1-3B-heretic-mlx-5bit) was converted to MLX format from [heretic-org/Nanbeige4.1-3B-heretic](https://huggingface.co/heretic-org/Nanbeige4.1-3B-heretic) using mlx-lm version **0.31.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("usermma/Nanbeige4.1-3B-heretic-mlx-5bit")

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)
```