---
pipeline_tag: text-generation
library_name: mlx
inference: false
tags:
- facebook
- meta
- llama
- llama-2
- mlx
license: llama2
---

# **Llama 2 7B Chat 4-bit**

Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 7B fine-tuned model, in `npz` format suitable for use in Apple's MLX framework.

Weights have been converted to `float16` from the original `bfloat16` type, because `numpy` is not compatible with `bfloat16` out of the box.

How to use with [MLX](https://github.com/ml-explore/mlx).

```bash
# Install mlx, mlx-examples, huggingface-cli
pip install mlx
pip install huggingface_hub hf_transfer
git clone https://github.com/ml-explore/mlx-examples.git

# Download model
export HF_HUB_ENABLE_HF_TRANSFER=1
huggingface-cli download --local-dir Llama-2-7b-chat-mlx/ mlx-community/Llama-2-7b-chat-4-bit

# Run example
python mlx-examples/llms/llama/llama.py --prompt "My name is " --model-path Llama-2-7b-chat-mlx/
```

Please, refer to the [original model card](https://huggingface.co/meta-llama/Llama-2-7b-chat) for details on Llama 2.

# Beware

This model's weights were converted to .npz, the old mlx format. You can use it to generate text but you cannot load it for fine tuning. [npz was adapted to the safetensor format](https://github.com/ml-explore/mlx-examples/issues/480#issuecomment-1960694170).
If you are looking for a quantized model, you can either use `mlx_lm.convert` on a hugging face model or find one that someone has already quantized and uploaded - but make sure to look in the Files and Versions tab for safetensors!