[LLM] AzureOpenAI 사용해보기

이경용 대표님 강의안

 

https://cafe.naver.com/aiclubcafe/1038

LLM - Azure OpenAI API

 

AuerOpenAI를 실행하여 AI와 관련된 Key, 모델 등을 설정한다.

 

이제 FastAPI로 배포해보자.

FastAPI 폴더의 main.py를 수정하자.

from fastapi import FastAPI, Form
from pydantic import BaseModel
from fastapi.middleware.cors import CORSMiddleware
import httpx
import json
from openai import AzureOpenAI

app = FastAPI()

# Azure OpenAI 인증 정보
azure_oai_endpoint = "https://azureopenai-flyai.openai.azure.com/" # 실제 endpoint 정보 입력해주자!
azure_oai_key = "KEY" # 실제 인증 Key 입력해주자!

# CORS 설정
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

client = AzureOpenAI(
        azure_endpoint = azure_oai_endpoint, 
        api_key=azure_oai_key,  
        api_version="2024-08-01-preview"
)

class RequestModel(BaseModel):
    model: str
    prompt: str

class Message(BaseModel):
    message: str

@app.post("/chat")
async def chat(message: Message):
    prompt = message.message
    # 요청할 URL
    url = "http://host.docker.internal:11434/api/generate"
    # 요청에 포함할 데이터
    data = {
        "model": "llama3.2-flyai-v1.0:latest",
        "prompt": prompt
    }
    # 요청 헤더
    headers = {
        "Content-Type": "application/json"
    }
    
    async with httpx.AsyncClient(timeout=20.0) as client:
        response = await client.post(url, headers=headers, data=json.dumps(data))

    if response.status_code == 200:
        # 개별 JSON 객체로 분할
        json_objects = response.content.decode().strip().split("\n")
        # 각 JSON 객체를 Python 사전으로 변환
        data = [json.loads(obj) for obj in json_objects]
        res_text = ''
        # 변환된 데이터 출력
        for item in data:
            print(item)
            res_text += item['response']
    else:
        res_text = f"Error Code : {response.status_code}"
    return {"reply": res_text}

@app.post("/openai")
async def chat(message: Message):
    prompt = message.message
    response = client.chat.completions.create(
        model="o1-mini",
        messages=[
            {"role": "user", "content": prompt}
        ]
    )
    generated_text = response.choices[0].message.content
    return {"reply": generated_text}

 

localhost:8000/docs 실행