Instructions to use AXCXEPT/phi-4-open-R1-Distill-EZOv1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AXCXEPT/phi-4-open-R1-Distill-EZOv1 with Transformers:

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

pipe = pipeline("text-generation", model="AXCXEPT/phi-4-open-R1-Distill-EZOv1", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AXCXEPT/phi-4-open-R1-Distill-EZOv1", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("AXCXEPT/phi-4-open-R1-Distill-EZOv1", trust_remote_code=True)
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 Settings

vLLM

How to use AXCXEPT/phi-4-open-R1-Distill-EZOv1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AXCXEPT/phi-4-open-R1-Distill-EZOv1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AXCXEPT/phi-4-open-R1-Distill-EZOv1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AXCXEPT/phi-4-open-R1-Distill-EZOv1

SGLang

How to use AXCXEPT/phi-4-open-R1-Distill-EZOv1 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 "AXCXEPT/phi-4-open-R1-Distill-EZOv1" \
    --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": "AXCXEPT/phi-4-open-R1-Distill-EZOv1",
		"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 "AXCXEPT/phi-4-open-R1-Distill-EZOv1" \
        --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": "AXCXEPT/phi-4-open-R1-Distill-EZOv1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AXCXEPT/phi-4-open-R1-Distill-EZOv1 with Docker Model Runner:
```
docker model run hf.co/AXCXEPT/phi-4-open-R1-Distill-EZOv1
```

AXCXEPT commited on Jan 27, 2025

Commit

0e807ef

verified ·

1 Parent(s): 19e4bd0

Update README.md

Browse files

Files changed (1) hide show

README.md +9 -3

README.md CHANGED Viewed

@@ -20,17 +20,19 @@ base_model:
 ## Model Details
 Deepseek-R1のDistill手法を模倣した、open-r1を採用して、phi-4モデルを Reasonerにしたモデルです。
 ## HOW TO USE
 ### Setup
 ```
 pip install --upgrade transformers accelerate datasets trl
 ```
 ### Predict(using AutoModelForCausalLM)
 ```python
 ```
 ### Predict(using vllm)
 #### Setup
 ```
@@ -61,4 +63,8 @@ completion = client.chat.completions.create(
 )
 print(completion.choices[0].message.content)
-```

 ## Model Details
 Deepseek-R1のDistill手法を模倣した、open-r1を採用して、phi-4モデルを Reasonerにしたモデルです。
+--------------------------------------
 ## HOW TO USE
+-------------------
 ### Setup
 ```
 pip install --upgrade transformers accelerate datasets trl
 ```
+-------------------
 ### Predict(using AutoModelForCausalLM)
 ```python
 ```
+-------------------
 ### Predict(using vllm)
 #### Setup
 ```
 )
 print(completion.choices[0].message.content)
+```
+### Special Thanks
+Phi-4 develop team, open-r1 team developer, deepseek team, thanks for your special technology and knowledge.