from openai import OpenAI
import datetime

# Modify OpenAI's API key and API base to use vLLM's API server.
openai_api_key = "EMPTY"
openai_api_base = "http://127.0.0.1:8080/v1"

client = OpenAI(
    api_key=openai_api_key,
    base_url=openai_api_base,
    timeout=600,
    max_retries=0
)
#print(datetime.datetime.now())
models = client.models.list()
model = models.data[0].id


count = 0
print("")
with open("lm_eval_gsm8k_questions.txt", "r", encoding='utf-8') as data_file:
    questions = data_file.readlines()
    for question in questions:
        print ("handling question ", count)
        count += 1
        messages = [{"role": "user", "content": question}]
        try:
            response = client.chat.completions.create(model=model, messages=messages)
            content = response.choices[0].message.content
            print("response: ", content)
            if (len(content) < 10) :
                print ("Met exception during test. Question: ", question)
                print ("Response: ", content)
                break
        except:
            print ("Met exception during test. Question: ", question)
            break