README.md

---
license: llama3.1
language:
- en
base_model: SicariusSicariiStuff/Assistant_Pepe_8B
datasets:
- SicariusSicariiStuff/UBW_Tapestries
widget:
- text: Assistant_Pepe_8B
  output:
    url: https://huggingface.co/SicariusSicariiStuff/Assistant_Pepe_8B/resolve/main/Images/Assistant_Pepe_8B.png
tags:
- mlx
---

# ailexleon/Assistant_Pepe_8B-mlx-4Bit

The Model [ailexleon/Assistant_Pepe_8B-mlx-4Bit](https://huggingface.co/ailexleon/Assistant_Pepe_8B-mlx-4Bit) was converted to MLX format from [SicariusSicariiStuff/Assistant_Pepe_8B](https://huggingface.co/SicariusSicariiStuff/Assistant_Pepe_8B) using mlx-lm version **0.29.1**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("ailexleon/Assistant_Pepe_8B-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)
```