大模型多模态调用

def duomotaiocr(): def local_image_to_data_uri(image_path: str) -> str | None: """本地图片转为 data URI 格式(适配 image_url 字段)""" try: # 获取图片后缀 ext=image_path.split('.')[-1].lower() if ext not in ('png', 'jpg', 'jpeg'): print("仅支持png/jpg/jpeg格式") return None with open(image_path, "rb") as img_file: base64_str=base64.b64encode(img_file.read()).decode("utf-8") # 关键：包装成 data URI 格式 return f"data:image/{ext};base64,{base64_str}" except Exception as e: print(f"图片编码失败: {e}") return None import os from pathlib import Path from openai import OpenAI import pandas as pd client = OpenAI( # 若没有配置环境变量，请用百炼API Key将下行替换为：api_key="sk-xxx" api_key=api_key, base_url="https://dashscope.aliyuncs.com/compatible-mode/v1", ) file_object = client.files.create(file=Path("./结算单.pdf"), purpose="file-extract") completion = client.chat.completions.create( model="qwen2.5-vl-72b-instruct", # 模型列表：https://help.aliyun.com/zh/model-studio/getting-started/models messages= [ { "role": "user", "content": [ { "type": "text", "text": "描述下图片里的内容" }, { "type": "image_url", # 需要注意，Base64，图像格式(即image/{format})需要与支持的图片列表中的Content Type保持一致。"f"是字符串格式化的方法。 # PNG图像： f"data:image/png;base64,{base64_image}" # JPEG图像： f"data:image/jpeg;base64,{base64_image}" # WEBP图像： f"data:image/webp;base64,{base64_image}" "image_url": { "url": f"{local_image_to_data_uri('1.png')}" } } ] } ] ) a=completion.choices[0].message.content print(a) with open('output.csv', mode='w', encoding='utf-8') as file: file.write(a) return 'result.xlsx'