通义千问VL

import osfrom openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
)

completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=[
        {            "role": "system",            "content": [{"type": "text", "text": "You are a helpful assistant."}],
        },
        {            "role": "user",            "content": [
                {                    "type": "image_url",                    "image_url": {                        "url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241022/emyrja/dog_and_girl.jpeg"
                    },
                },
                {"type": "text", "text": "图中描绘的是什么景象?"},
            ],
        },
    ],
)print(completion.choices[0].message.content)

from openai import OpenAIimport os

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=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1")
messages = [
    {        "role": "system",        "content": [{"type": "text", "text": "You are a helpful assistant."}]},
    {        "role": "user",        "content": [
            {                "type": "image_url",                "image_url": {                    "url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241022/emyrja/dog_and_girl.jpeg"
                },
            },
            {"type": "text", "text": "图中描绘的是什么景象？"},
        ],
    }
]
completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=messages,
    )print(f"第一轮输出：{completion.choices[0].message.content}")
assistant_message = completion.choices[0].message
messages.append(assistant_message.model_dump())
messages.append({        "role": "user",        "content": [
        {            "type": "text",            "text": "做一首诗描述这个场景"
        }
        ]
    })
completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=messages,
    )print(f"第二轮输出：{completion.choices[0].message.content}")

from openai import OpenAIimport os

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=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
)

completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=[
	{"role": "system",         "content": [{"type":"text","text": "You are a helpful assistant."}]},
        {"role": "user",         "content": [{"type": "image_url",                    "image_url": {"url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241022/emyrja/dog_and_girl.jpeg"},},
                    {"type": "text", "text": "图中描绘的是什么景象？"}]}],
    stream=True)
full_content = ""print("流式输出内容为：")for chunk in completion:    if chunk.choices[0].delta.content is None:        continue
    full_content += chunk.choices[0].delta.content    print(chunk.choices[0].delta.content)print(f"完整内容为：{full_content}")

from openai import OpenAIimport osimport base64#  base 64 编码格式def encode_image(image_path):    with open(image_path, "rb") as image_file:        return base64.b64encode(image_file.read()).decode("utf-8")# 将xxxx/test.png替换为你本地图像的绝对路径base64_image = encode_image("xxx/test.png")
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=os.getenv('DASHSCOPE_API_KEY'),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
)
completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=[
    	{    	    "role": "system",            "content": [{"type":"text","text": "You are a helpful assistant."}]},
        {            "role": "user",            "content": [
                {                    "type": "image_url",                    # 需要注意，传入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"data:image/png;base64,{base64_image}"}, 
                },
                {"type": "text", "text": "图中描绘的是什么景象?"},
            ],
        }
    ],
)print(completion.choices[0].message.content)

import osfrom openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
)
completion = client.chat.completions.create(
    model="qwen-vl-max-latest",
    messages=[
       {"role":"system","content":[{"type": "text", "text": "You are a helpful assistant."}]},
       {"role": "user","content": [           # 第一张图像链接，如果传入本地文件，请将url的值替换为图像的Base64编码格式
           {"type": "image_url","image_url": {"url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20241022/emyrja/dog_and_girl.jpeg"},},           # 第二张图像链接，如果传入本地文件，请将url的值替换为图像的Base64编码格式
           {"type": "image_url","image_url": {"url": "https://dashscope.oss-cn-beijing.aliyuncs.com/images/tiger.png"},},
           {"type": "text", "text": "这些图描绘了什么内容？"},
            ],
        }
    ],
)print(completion.choices[0].message.content)