README.md

---
language:
- code
license: llama2
tags:
- llama-2
- mlx
pipeline_tag: text-generation
---

![Alt text](https://cdn.discordapp.com/attachments/1064373193982361601/1202164612645265418/codellama70bbase.png?ex=65cc760a&is=65ba010a&hm=d8db3259380c5faa567b20614af3c1c203a459fa7fbf3e01221bb80d9a95e246&)

# mlx-community/CodeLlama-70b-hf-4bit-MLX
This model was converted to MLX format from [`codellama/CodeLlama-70b-hf`]().
Refer to the [original model card](https://huggingface.co/codellama/CodeLlama-70b-hf) for more details on the model.

This can be used as base for additional fine-tuning experiments.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/CodeLlama-70b-hf-4bit-MLX")
response = generate(model, tokenizer, prompt="hello", verbose=True)
```