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
```

Phi-4 with Tools

#28

by tahafatih - opened Jan 15, 2025

Discussion

tahafatih

Jan 15, 2025

Is it possible to use the Phi-4 model with Tools?

RecViking

Jan 15, 2025

I've been using Phi-4 with json style and python tool calling (in smolagents) agents. I was loading it with llama.cpp, so I had to write my own "model" to support it, but phi-4 is my favorite 'smaller' tool calling model I've used so far.

mayonesa

Feb 21, 2025

@RecViking please elaborate on how you were able to use Phi-4 w/ tool-calling

RecViking

Feb 25, 2025

@mayonesa - I created a new "model" within the system so I could manually manipulate the grammars passed to the model. With some of the recent updates, the openai compatibility has improved and allows for additional parameters. You should be able to self host using something like llama.cpp and set up grammars for tool calling using the openai "model" (wrapper for openai library).

RobCzikkel

Feb 26, 2025

@RecViking please elaborate on how you were able to use Phi-4 w/ tool-calling

with smolagents, phi-4 also works using a TGI server. just set up an endpoint with TGI, use the OpenAIServerModel from smolagents and you are pretty much good to go

nbaughman

Aug 31, 2025

@mayonesa could you potentially link your implementation in a gist? I'm trying to get it working in vLLM/Pydantic AI but I'm having a hell of a time trying to write a tool call parser for it

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