jy-hxy's picture
bee386f048e4bb26a89614be34d9a61de73f2184d8066d0775f544a1414cd74c
56094a1 verified
|
raw
history blame
844 Bytes
---
license: gpl-3.0
tags:
- mlx
base_model: CausalLM/34b-beta
---
# mlx-community/CausalLM-34b-beta-4bit
The Model [mlx-community/CausalLM-34b-beta-4bit](https://huggingface.co/mlx-community/CausalLM-34b-beta-4bit) was
converted to MLX format from [CausalLM/34b-beta](https://huggingface.co/CausalLM/34b-beta)
using mlx-lm version **0.20.4**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/CausalLM-34b-beta-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)
```