Image generation - async submit (OpenAI Images compatible)

POST

/v1/images/generations

curl --request POST \
  --url https://coder.api.visioncoder.cn/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "model": "gpt-image-2", "prompt": "一只戴着宇航员头盔的橘猫，赛博朋克风格", "n": 1, "size": "1:1", "resolution": "1k", "image_urls": [ "example" ] }'

• Production

Authorizations

• BearerAuth

Request Body^required

Media typeapplication/json

CreateImagesGenerationReq

OpenAI Images-compatible generation request. The task runs asynchronously: this endpoint returns a task ID, and the result is fetched via GET /v1/tasks/{task_id}.

object

model

required

Image model ID, e.g. gpt-image-2. Call GET /v1/models for the live list.

string

Example

gpt-image-2

prompt

required

Text description of the image to generate. Chinese and English are both supported.

string

Example

一只戴着宇航员头盔的橘猫，赛博朋克风格

n

Number of images to generate (1-10).

integer

default: 1 >= 1 <= 10

size

Aspect ratio such as 1:1, 16:9, auto; pixel sizes like 1024x1024 are also accepted.

string

Example

1:1

resolution

Output resolution tier.

string

default: 1k

Allowed values: 1k 2k 4k

image_urls

Reference images (URL or base64). Providing this switches to image-to-image mode; up to 14 images.

Array<string>

<= 14 items

Responses

200

Task submitted. Poll GET /v1/tasks/{task_id} with the returned id to fetch the result.

Media typeapplication/json

CreateImagesGenerationResp

object

code

required

Status code; 200 means the task was submitted successfully.

integer

data

required

ImagesTaskData

object

id

required

Task ID, used to poll GET /v1/tasks/{task_id}.

string

status

required

Task status; submitted right after submission.

string

Example

{
  "code": 200,
  "data": {
    "id": "task_01HZY3K8Q2W7X9V4",
    "status": "submitted"
  }
}