2025年GPT-4o API完全指南：接口调用、多模态功能与最佳实践

# Python环境安装openai库
pip install openai

# 或者使用requests库也可以进行API调用
pip install requests

# 使用官方OpenAI库调用
import openai

# 设置API密钥和基础URL
client = openai.OpenAI(
    api_key="sk-xxx",  # 替换为您的API密钥
    # 使用官方API时，不需要设置base_url
    # 使用laozhang.ai中转API时，设置为：
    # base_url="https://api.laozhang.ai/v1"
)

# 调用GPT-4o模型
response = client.chat.completions.create(
    model="gpt-4o",  # 指定使用GPT-4o模型
    messages=[
        {"role": "system", "content": "你是一个专业的AI助手，擅长解答技术问题。"},
        {"role": "user", "content": "请解释什么是多模态AI模型以及其主要应用场景。"}
    ],
    temperature=0.7,  # 控制创造性，0-1之间，越高创造性越强
    max_tokens=1000   # 限制回复长度
)

# 输出模型回复
print(response.choices[0].message.content)

# 使用requests库调用laozhang.ai中转API
import requests
import json

# API配置
api_key = "lz_xxx"  # 替换为您的laozhang.ai API密钥
api_url = "https://api.laozhang.ai/v1/chat/completions"

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

# 请求数据
data = {
    "model": "gpt-4o",
    "messages": [
        {"role": "system", "content": "你是一个专业的AI助手，擅长解答技术问题。"},
        {"role": "user", "content": "请解释什么是多模态AI模型以及其主要应用场景。"}
    ],
    "temperature": 0.7,
    "max_tokens": 1000
}

# 发送请求
response = requests.post(api_url, headers=headers, json=data)
result = response.json()

# 获取模型回复
if "choices" in result and len(result["choices"]) > 0:
    print(result["choices"][0]["message"]["content"])
else:
    print(f"请求失败: {result}")

# 图像理解示例 - 官方SDK
import openai
import base64

# 读取并编码图像
def encode_image(image_path):
    with open(image_path, "rb") as image_file:
        return base64.b64encode(image_file.read()).decode('utf-8')

# 获取base64编码的图像
base64_image = encode_image("your_image.jpg")

# 初始化OpenAI客户端
client = openai.OpenAI(api_key="sk-xxx")  # 替换为您的API密钥

# 创建带图像的请求
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "这张图片里有什么内容？详细描述一下。"},
                {"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"}}
            ]
        }
    ],
    max_tokens=1000
)

print(response.choices[0].message.content)

# 图像生成示例 - laozhang.ai API
import requests
import json

api_key = "lz_xxx"  # 替换为您的API密钥
api_url = "https://api.laozhang.ai/v1/images/generations"

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

data = {
    "model": "gpt-4o",  # 使用GPT-4o模型生成图像
    "prompt": "一只戴着太阳镜的猫咪，在海滩上冲浪，逼真风格，4K高清",
    "n": 1,
    "size": "1024x1024",
    "quality": "standard",
    "style": "vivid"  # 可选natural或vivid
}

response = requests.post(api_url, headers=headers, json=data)
result = response.json()

if "data" in result and len(result["data"]) > 0:
    image_url = result["data"][0]["url"]
    print(f"生成的图像URL: {image_url}")
else:
    print(f"生成失败: {result}")

# 文本转语音示例 - 官方SDK
import openai

client = openai.OpenAI(api_key="sk-xxx")  # 替换为您的API密钥

response = client.audio.speech.create(
    model="gpt-4o",  # 注意：实际使用时请确认最新支持的模型名称
    voice="alloy",  # 可选voice: alloy, echo, fable, onyx, nova, shimmer
    input="你好，我是GPT-4o模型。我可以提供多模态智能助手服务，包括文本处理、图像分析和生成等功能。"
)

# 保存为音频文件
response.stream_to_file("output.mp3")

# 流式输出示例 - 官方SDK
import openai

client = openai.OpenAI(api_key="sk-xxx")  # 替换为您的API密钥

# 启用流式输出
stream = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "你是一个专业的AI助手。"},
        {"role": "user", "content": "请写一篇关于人工智能发展历史的短文。"}
    ],
    stream=True  # 开启流式输出
)

# 逐步获取并输出内容
for chunk in stream:
    if chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end="")

# 文档分析示例
import openai
import base64

def encode_image(image_path):
    with open(image_path, "rb") as image_file:
        return base64.b64encode(image_file.read()).decode('utf-8')

# 获取base64编码的文档图像
base64_document = encode_image("scanned_document.jpg")

client = openai.OpenAI(api_key="sk-xxx")  # 替换为您的API密钥

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "这份文档包含哪些关键信息？请提取其中的表格数据并整理成JSON格式。"},
                {"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64_document}"}}
            ]
        }
    ],
    max_tokens=1500
)

print(response.choices[0].message.content)