Llama-2-7b-chat-4-bit

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.

# 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 for details on Llama 2.