Update README.md

README.md

@@ -20,3 +20,45 @@ 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
+```
+pip install vllm 
+vllm serve AXCXEPT/phi-4-open-R1-Distill-EZOv1
+```
+#### Predict
+```
+from openai import OpenAI
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="token-abc123",
+)
+
+prompt = f"""
+There is a question concerning the age of a father and his child. The conditions are as follows
+
+The father's current age is 3 times the son's current age.
+Six years ago, the father's age was five times the son's age.
+Find the ages of the father and the son.
+"""
+completion = client.chat.completions.create(
+  model="AXCXEPT/EZO-phi-4-openr1-v1_917",
+  messages=[
+    {"role": "system", "content": "You are an excellent AI. Please answer carefully and thoughtfully, in the same language as the instructions."}
+    {"role": "user", "content": prompt}
+  ]
+)
+
+print(completion.choices[0].message.content)
+```