---
license: other
license_name: katanemo-research
license_link: https://huggingface.co/katanemo/Arch-Agent-32B/blob/main/LICENSE
base_model: katanemo/Arch-Agent-32B
language:
- en
pipeline_tag: text-generation
library_name: transformers
tags:
- mlx
---

# huangang/Arch-Agent-32B-mlx-4Bit

The Model [huangang/Arch-Agent-32B-mlx-4Bit](https://huggingface.co/huangang/Arch-Agent-32B-mlx-4Bit) was converted to MLX format from [katanemo/Arch-Agent-32B](https://huggingface.co/katanemo/Arch-Agent-32B) using mlx-lm version **0.22.3**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("huangang/Arch-Agent-32B-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)
```