ChatGPT 4o图像API全指南：2025年最新功能与实战应用

# 图像理解示例代码
import requests
import base64
import json

# API端点（通过laozhang.ai中转服务访问）
API_URL = "https://api.laozhang.ai/v1/chat/completions"
API_KEY = "your_api_key_here" # 在laozhang.ai注册获取

# 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')

# 图像路径
image_path = "path/to/your/image.jpg"
base64_image = encode_image(image_path)

# 构建请求
headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {API_KEY}"
}

payload = {
    "model": "gpt-4o",
    "messages": [
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "这张图片展示了什么内容？请详细描述。"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": f"data:image/jpeg;base64,{base64_image}"
                    }
                }
            ]
        }
    ],
    "max_tokens": 500
}

# 发送请求
response = requests.post(API_URL, headers=headers, json=payload)
print(json.dumps(response.json(), indent=4))

# 图像生成示例代码
import requests
import json
import os

# API端点（通过laozhang.ai中转服务访问）
API_URL = "https://api.laozhang.ai/v1/chat/completions"
API_KEY = "your_api_key_here" # 在laozhang.ai注册获取

# 构建请求
headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {API_KEY}"
}

payload = {
    "model": "gpt-4o",
    "messages": [
        {
            "role": "system",
            "content": "你是一个专业的图像生成助手。请根据用户的描述生成高质量图像。"
        },
        {
            "role": "user",
            "content": "生成一张未来智能城市的图像，包含飞行汽车、智能建筑和绿色能源设施。风格为明亮的科技感。"
        }
    ],
    "max_tokens": 1000,
    "modalities": ["image"]  # 启用图像生成
}

# 发送请求
response = requests.post(API_URL, headers=headers, json=payload)
result = response.json()

# 处理返回的图像URL
if "image_url" in str(result):
    for message in result.get("choices", [{}])[0].get("message", {}).get("content", []):
        if isinstance(message, dict) and message.get("type") == "image_url":
            image_url = message.get("image_url", {}).get("url")
            if image_url:
                print(f"生成的图像URL: {image_url}")
                # 此处可添加下载图像的代码
else:
    print("响应中没有找到图像URL")
    print(json.dumps(result, indent=4))

# 标准API请求结构
curl https://api.laozhang.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $API_KEY" \
  -d '{
    "model": "gpt-4o",
    "stream": false,
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"} 
    ]
  }'

# 优化的提示词模板
prompt_template = """
主体：{subject_description}
场景：{setting_description}
光照：{lighting_description}
风格：{style_description}
视角：{perspective}
技术参数：{technical_details}
"""

# 实例
optimized_prompt = prompt_template.format(
    subject_description="一位身穿未来风格深蓝色商务套装的亚洲女性企业家",
    setting_description="现代化科技办公室，大型落地窗，城市天际线背景",
    lighting_description="明亮的自然光从左侧窗户照入，柔和的室内灯光",
    style_description="逼真的商业摄影风格，清晰锐利，专业色彩校准",
    perspective="中距离半身像，略微仰视角度",
    technical_details="高分辨率，浅景深，f/2.8光圈，自然肤色"
)

# 图像预处理优化示例
from PIL import Image
import io
import base64

def optimize_image_for_api(image_path, max_size=(1024, 1024), quality=85):
    """优化图像以减少API调用成本"""
    # 打开图像
    img = Image.open(image_path)
    
    # 调整大小，保持宽高比
    img.thumbnail(max_size, Image.LANCZOS)
    
    # 转换为JPEG并压缩
    buffer = io.BytesIO()
    img.convert('RGB').save(buffer, format="JPEG", quality=quality)
    buffer.seek(0)
    
    # 转为base64
    return base64.b64encode(buffer.read()).decode('utf-8')

# 使用优化后的图像
optimized_base64 = optimize_image_for_api("original_image.jpg")

# 智能重试机制
import time
import random

def call_api_with_retry(payload, max_retries=5, base_delay=1):
    """实现智能重试机制的API调用"""
    for attempt in range(max_retries):
        try:
            response = requests.post(API_URL, headers=headers, json=payload, timeout=30)
            
            # 处理成功响应
            if response.status_code == 200:
                return response.json()
                
            # 处理速率限制
            if response.status_code == 429:
                # 获取重试间隔（如果有）
                retry_after = int(response.headers.get('Retry-After', base_delay))
                # 增加随机抖动避免请求同步
                jitter = random.uniform(0.1, 0.5)
                sleep_time = retry_after + jitter
                print(f"速率限制，等待 {sleep_time:.2f} 秒后重试...")
                time.sleep(sleep_time)
                continue
                
            # 其他错误
            response.raise_for_status()
            
        except requests.exceptions.RequestException as e:
            # 计算指数退避延迟
            delay = base_delay * (2 ** attempt) + random.uniform(0, 1)
            print(f"请求失败: {e}. 等待 {delay:.2f} 秒后重试 (尝试 {attempt+1}/{max_retries})")
            
            if attempt == max_retries - 1:
                raise Exception(f"达到最大重试次数: {e}")
                
            time.sleep(delay)
    
    raise Exception("未知错误：所有重试均失败")