文档 / 示例 / 文本生成图像

文本生成图像

使用 POST /v1/images/generations 从文本提示词生成图像。一次调用，响应中即返回图像 URL。

#Python

text_to_image.pypython

import requests, os

response = requests.post(
    "https://relayrouter.io/v1/images/generations",
    headers={
        "Authorization": f"Bearer {os.environ['RELAYROUTER_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": "gpt-image-1",
        "prompt": "a cinematic shot of an astronaut floating above Earth at sunrise, photorealistic",
        "n": 1,
    },
)
response.raise_for_status()
print(response.json()["data"][0]["url"])

#Node.js / TypeScript

textToImage.tstypescript

const res = await fetch("https://relayrouter.io/v1/images/generations", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.RELAYROUTER_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "gpt-image-1",
    prompt: "a cinematic shot of an astronaut floating above Earth at sunrise, photorealistic",
    n: 1,
  }),
});
const { data } = await res.json();
console.log("Image URL:", data[0].url);

#cURL

terminalbash

curl https://relayrouter.io/v1/images/generations \
  -H "Authorization: Bearer $RELAYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "a cinematic shot of an astronaut floating above Earth at sunrise, photorealistic",
    "n": 1
  }'

#参数

参数	类型	说明
`model`	string	图像模型 ID（必填）
`prompt`	string	图像的文本描述（必填）
`n`	integer	要生成的图像数量。默认值：1
`size`	string	可选的输出尺寸，取决于具体模型（例如 "1024x1024"）

图像模型 ID 可通过 GET /v1/models 或模型页面查找。完整的端点细节请参阅 Generate Image 参考。

←CLI vLLM→