inference/vllm_ascend/examples/quick_start.py

import json
import base64
import os
import requests
import json
from PIL import Image, ImageDraw, ImageFont
import ast
import re
def encode_image_to_base64(img_path):
    """
    :param img_path: image path
    :return: Base64 encoding
    """
    try:
        with open(img_path, 'rb') as img_file:
            img_data = img_file.read()
        base64_str = base64.b64encode(img_data).decode('utf-8')
        return base64_str
    except Exception as e:
        print(f"Encoding Error: {e}")
        return None

def infer_image_with_api(image_path, prompt):
    url = "http://127.0.0.1:8000/v1/chat/completions"
    base64_image = encode_image_to_base64(image_path)
    payload = json.dumps({
      "messages": [
          {
                "role": "system",
                "content": [
                    {"type": "text", "text": "你是华为公司开发的多模态大模型，名字是openPangu-VL-7B。你能够处理文本和视觉模态的输入，并给出文本输出。"},     
                ]
          },
          {
                "role": "user",
                "content": [
                    {"type": "image_url", "image_url": {"url": f"data:image/jpg;base64,{base64_image }"}},
                    {"type": "text", "text": prompt},     
                ]
          }
      ],    
      "model": "pangu_vl",
      "max_tokens": 2048,
      "temperature": 0,
      "stream": False,

    })

    headers = {
      'Content-Type': 'application/json'
    }

    response = requests.request("POST", url, headers=headers, data=payload)
    return json.loads(response.text)["choices"][0]["message"]["content"]


def infer_message_with_api(prompt):
    url = "http://127.0.0.1:8000/v1/chat/completions"
    payload = json.dumps({
      "messages": [
          {
                "role": "system",
                "content": [
                    {"type": "text", "text": "你是华为公司开发的多模态大模型，名字是openPangu-VL-7B。你能够处理文本和视觉模态的输入，并给出文本输出。"},     
                ]
          },
          {
                "role": "user",
                "content": [
                    {"type": "text", "text": prompt},     
                ]
          }
      ],    
      "model": "pangu_vl",
      "max_tokens": 2048,
      "temperature": 0,
      "stream": False,

    })

    headers = {
      'Content-Type': 'application/json'
    }

    response = requests.request("POST", url, headers=headers, data=payload)
    return json.loads(response.text)["choices"][0]["message"]["content"]

res = infer_message_with_api("你好，你是谁？")
print(res)