Instructions to use microsoft/phi-4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/phi-4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="microsoft/phi-4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/phi-4")
model = AutoModelForCausalLM.from_pretrained("microsoft/phi-4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use microsoft/phi-4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/phi-4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/phi-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/microsoft/phi-4

SGLang

How to use microsoft/phi-4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "microsoft/phi-4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/phi-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "microsoft/phi-4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/phi-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use microsoft/phi-4 with Docker Model Runner:
```
docker model run hf.co/microsoft/phi-4
```

why not release the 7b model?

#18

by breadlicker45 - opened Jan 9, 2025

Discussion

breadlicker45

Jan 9, 2025

•

edited Jan 9, 2025

in the phi4 paper it shows that there is a 7b model

is there a reason besides "safety" it isn't released?

JLouisBiz

Jan 9, 2025

I would like to have smaller one too, I am using phi-3.5-mini-instruct with success, though would like some upgrade to run on my 16 GB RAM and 4 GB GPU

Mizule

Jan 9, 2025

I would like to have smaller one too, I am using phi-3.5-mini-instruct with success, though would like some upgrade to run on my 16 GB RAM and 4 GB GPU

try the unsloth 4 bit dynamic quant, it gets nearly the same performance as 16 bit and fits in under 15GB

JLouisBiz

Jan 9, 2025

Thank you, how would I unsloth, how do I do it? Which command? I know how to use llama-quantize command, but please help me with specific one.

Mizule

Jan 10, 2025

Thank you, how would I unsloth, how do I do it? Which command? I know how to use llama-quantize command, but please help me with specific one.

you can get the already quantized model here: https://huggingface.co/unsloth/phi-4-unsloth-bnb-4bit

it also links a colab notebook you can use for inference and finetuning of it, I assume all you have to do is change the model its using and lower the batch size.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment