Phi-4-mini-instruct-mlx-fp16/README.md

language, library_name, license, license_link, pipeline_tag, tags, widget, base_model

language	library_name	license	license_link	pipeline_tag	tags	widget	base_model
multilingual ar zh cs da nl en fi fr de he hu it ja ko no pl pt ru es sv th tr uk	transformers	mit	https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/LICENSE	text-generation	nlp code mlx mlx-my-repo	messages role content user Can you provide ways to eat combinations of bananas and dragonfruits?	microsoft/Phi-4-mini-instruct

pa5haw/Phi-4-mini-instruct-mlx-fp16

The Model pa5haw/Phi-4-mini-instruct-mlx-fp16 was converted to MLX format from microsoft/Phi-4-mini-instruct using mlx-lm version 0.31.2.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("pa5haw/Phi-4-mini-instruct-mlx-fp16")

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)