app.py

import base64
import requests
from io import BytesIO
from PIL import Image
import gradio as gr

def encode_image(img):
    buffered = BytesIO()
    img.save(buffered, format="PNG")
    encoded_string = base64.b64encode(buffered.getvalue()).decode("utf-8")
    return encoded_string

def chat_with_pixtral(uploaded_file, user_question):
    if uploaded_file is not None:
        # uploaded_file은 이미 PIL 이미지 객체입니다.
        base64_img = encode_image(uploaded_file)

        api = "https://api.hyperbolic.xyz/v1/chat/completions"
        api_key = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJyZzMyNzAyNEBnbWFpbC5jb20ifQ._frFve-BYZdb0Qo6FIj6xcDcxpY-6QlC2O-ToQxBjkc"  # 여기에 API 키를 입력하세요

        headers = {
            "Content-Type": "application/json",
            "Authorization": f"Bearer {api_key}",
        }

        payload = {
            "messages": [
                {
                    "role": "user",
                    "content": [
                        {"type": "text", "text": user_question},  # 사용자가 입력한 질문
                        {
                            "type": "image_url",
                            "image_url": {"url": f"data:image/jpeg;base64,{base64_img}"},
                        },
                    ],
                }
            ],
            "model": "mistralai/Pixtral-12B-2409",
            "max_tokens": 2048,
            "temperature": 0.7,
            "top_p": 0.9,
        }

        response = requests.post(api, headers=headers, json=payload)

        # API 응답 확인
        if response.status_code == 200:
            response_data = response.json()
            if 'choices' in response_data:
                assistant_response = response_data['choices'][0]['message']['content']
            else:
                assistant_response = "응답 형식이 올바르지 않습니다."
        else:
            assistant_response = f"API 요청 실패: {response.status_code} - {response.text}"

        return assistant_response
    return "이미지를 업로드하고 질문을 입력하세요."

# Gradio 인터페이스 설정
iface = gr.Interface(
    fn=chat_with_pixtral,
    inputs=[
        gr.Image(type="pil", label="이미지를 업로드하세요"),
        gr.Textbox(label="질문을 입력하세요")
    ],
    outputs="text",
    title="Pixtral Image Chat",
    description="이미지를 업로드하고 질문을 입력하여 Pixtral과 대화하세요."
)

iface.launch(share=True)