Generate image ​

POST https://openqiao.com/api/v1/images/generations

The Generate image endpoint is compatible with OpenAI's Create image endpoint and creates images from prompts.

The parameters below cover what different models may support. Support varies by model. 开桥 passes through the OpenAI Images API protocol; when calling 开桥, set the Base URL to https://openqiao.com/api/v1 and authenticate with your 开桥 API Key.

Request headers ​

Authorization string * ​

Bearer Token authentication, formatted as Bearer $OPENQIAO_API_KEY.

Content-Type string * ​

The request content type. Defaults to application/json.

Request body ​

prompt string * ​

Text description of the image to generate. GPT image models support a maximum length of 32,000 characters.

background string Optional ​

Sets the background transparency of the generated image. This parameter is only supported by GPT image models. Must be one of transparent, opaque, or auto; defaults to auto. With auto, the model automatically determines the most suitable background for the image.

If set to transparent, the output format must support transparency, so output_format should be set to png (default) or webp.

Possible values:

• transparent
• opaque
• auto

model string Optional ​

The model to use for image generation. Visit the 开桥 model catalog to check the currently available models.

moderation string Optional ​

Controls the content moderation level for GPT image model generation. Must be one of low or auto (default).

Possible values:

• low
• auto

n number Optional ​

The number of images to generate. Must be between 1 and 10.

output_compression number Optional ​

The compression level for generated images (0-100%). This parameter is only supported by GPT image models with webp or jpeg output formats, and defaults to 100.

output_format string Optional ​

The format used to return generated images. This parameter is only supported by GPT image models. Must be one of png, jpeg, or webp.

Possible values:

• png
• jpeg
• webp

partial_images number Optional ​

The number of partial images to generate. This parameter is used for streaming responses that return partial images. The value must be between 0 and 3. When set to 0, the response returns a single image in one streaming event.

If the complete image is generated faster, the final image may be sent before all requested partial images have been produced.

quality string Optional ​

The quality of the generated image. 开桥 currently supports GPT image model quality parameters for OpenAI image generation models.

• auto (default) automatically selects the best quality for the given model.
• high, medium, and low are supported by GPT image models.

Possible values:

• low
• medium
• high
• auto

response_format string Optional ​

The return format for the generated image. This parameter does not apply to the GPT image models currently supported by 开桥; GPT image models always return base64-encoded images.

Possible values:

• url
• b64_json

size string Optional ​

The size of the generated image. For gpt-image-2 and gpt-image-2-2026-04-21, arbitrary WIDTHxHEIGHT strings are supported, such as 1536x864. Both width and height must be divisible by 16, and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above 2560x1440 are experimental, with a maximum supported resolution of 3840x2160. Requested sizes must also satisfy the model's current pixel and side-length limits.

gpt-image-1.5 supports the standard sizes 1024x1024, 1536x1024, and 1024x1536; models that support automatic sizing may use auto.

Common possible values:

• auto
• 1024x1024
• 1536x1024
• 1024x1536

stream boolean Optional ​

Generate the image in streaming mode. Defaults to false. This parameter is only supported by GPT image models.

style string Optional ​

The style of the generated image. This parameter does not apply to the OpenAI image generation models currently supported by 开桥.

Possible values:

• vivid
• natural

user string Optional ​

A unique identifier representing your end user, which can help with monitoring and abuse detection.

Response ​

Returns the response object from the image generation endpoint.

created number ​

The Unix timestamp (in seconds) when the image was created.

background string Optional ​

The background parameter used during image generation. May be transparent or opaque.

data array Optional ​

The list of generated images.

b64_json string Optional ​

The base64-encoded JSON for the generated image. GPT image models currently supported by 开桥 return this field by default.

revised_prompt string Optional ​

Returned by some models. Represents the rewritten prompt actually used to generate the image.

url string Optional ​

May contain a generated image URL for models that return URLs. GPT image models currently supported by 开桥 usually return b64_json.

output_format string Optional ​

The output format of the generated image. May be png, webp, or jpeg.

quality string Optional ​

The generated image quality. May be low, medium, or high.

size string Optional ​

The generated image size. May be 1024x1024, 1024x1536, or 1536x1024.

usage object Optional ​

Token usage information for GPT image model image generation.

• input_tokens number: The number of image and text tokens in the input prompt.
• input_tokens_details object: Input token details.
  • image_tokens number: The number of image tokens in the input prompt.
  • text_tokens number: The number of text tokens in the input prompt.
• output_tokens number: The number of output tokens generated by the model.
• total_tokens number: The total number of tokens used for this image generation.
• output_tokens_details object Optional: Output token details.
  • image_tokens number: The number of generated image output tokens.
  • text_tokens number: The number of generated text output tokens.

curl https://openqiao.com/api/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENQIAO_API_KEY" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A cute baby sea otter",
    "n": 1,
    "size": "1024x1024"
  }'

import OpenAI from "openai";
import { writeFile } from "fs/promises";

const client = new OpenAI({
  baseURL: "https://openqiao.com/api/v1", 
  apiKey: process.env.OPENQIAO_API_KEY, 
});

const img = await client.images.generate({
  model: "gpt-image-2",
  prompt: "A cute baby sea otter",
  n: 1,
  size: "1024x1024",
});

const imageBuffer = Buffer.from(img.data[0].b64_json, "base64");
await writeFile("output.png", imageBuffer);

import base64
from openai import OpenAI

client = OpenAI(
    base_url="https://openqiao.com/api/v1",  
    api_key="<OPENQIAO_API_KEY>",  
)

img = client.images.generate(
    model="gpt-image-2",
    prompt="A cute baby sea otter",
    n=1,
    size="1024x1024"
)

image_bytes = base64.b64decode(img.data[0].b64_json)
with open("output.png", "wb") as f:
    f.write(image_bytes)

{
  "created": 1713833628,
  "data": [
    {
      "b64_json": "..."
    }
  ],
  "usage": {
    "total_tokens": 100,
    "input_tokens": 50,
    "output_tokens": 50,
    "input_tokens_details": {
      "text_tokens": 10,
      "image_tokens": 40
    }
  }
}