Phi-3.5-mini-instruct-mlx-int4

This is a quantized INT4 model based on Apple MLX Framework Phi-3.5-mini-instruct. You can deploy it on Apple Silicon devices (M1,M2,M3).

Note: This is unoffical version,just for test and dev.

Installation

pip install -U mlx-lm 

Conversion

python -m mlx_lm.convert --hf-path microsoft/Phi-3.5-mini-instruct -q

Samples

from mlx_lm import load, generate

model, tokenizer = load("Your Phi-3.5-instruct-mlx-int4 Path")

prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": "能介绍一下 iOS 吗？"}],
    tokenize=False,
    add_generation_prompt=True,
)

response = generate(model, tokenizer, prompt=prompt,max_tokens=1024, verbose=True)