Create GPT Image 2.6 task

POST

images

curl --request POST \
  --url https://sigrunai.com/v1/images \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "gpt-image-2",
  "prompt": "Minimal product still life, white background, soft light",
  "quality": "medium",
  "n": 1,
  "ratio": "1:1"
}
'

{
  "code": 20000,
  "msg": "<string>",
  "data": {
    "task_id": "<string>",
    "status": "<string>",
    "progress": 123,
    "model": "<string>",
    "created_at": 123,
    "completed_at": 123,
    "cost": 123,
    "error_message": "<string>",
    "result": {
      "image_url": "<string>",
      "video_url": "<string>"
    }
  }
}

Create an async GPT Image 2.6 task. Poll with Query image task and read result.image_url when completed. For GPT Image 2, see Create GPT Image 2 task. For Nano Banana, see Create Nano Banana task.

Models

`model`	Notes	Billing
`gpt-image-2.6-flare`	Faster	`per_task`
`gpt-image-2.6-sunburst`	Finer detail	`per_task`

Each model has its own per-task price. quality changes the render, not the price. Check the model catalog and cost on the create response for the current unit price. You can also put quality in the model id, for example gpt-image-2.6-sunburst-max. That only sets quality. Billing still uses gpt-image-2.6-sunburst.

Request body

Field	Type	Required	Description
`model`	string	yes	See the table above
`prompt`	string	yes	1–32000 characters
`quality`	string	no	`auto` (default), `low`, `medium`, `high`, `xhigh`, `max`
`n`	int	no	Only 1. Omitted requests bill one image
`ratio`	string	no	`auto`, `1:1`, `3:2`, `2:3`, `4:3`, `3:4`, `5:4`, `4:5`, `16:9`, `9:16`, `21:9`
`size`	string	no	Alias of `ratio`. Pixel sizes such as `1024x1024` are rejected
`resolution`	string	no	Omit it, or send `1k`
`output_format`	string	no	`png`, `jpeg`, `webp`, `svg`, `gif`. `jpg` is treated as `jpeg`
`images`	string[]	no	Public reference image URLs, max 16
`smart_queue`	bool	no	Default `false`. When `true`, a very low recent success rate keeps the task `queued` until recovery (up to about 2 hours). See Smart queue

svg and gif cannot include reference images. For gif, omit size / ratio or send auto.

Example

Text to image

curl -X POST 'https://sigrunai.com/v1/images' \
  -H 'Authorization: Bearer sk_xxx' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "gpt-image-2.6-flare",
    "prompt": "Minimal product still life, white background, soft light",
    "quality": "medium",
    "n": 1,
    "ratio": "1:1",
    "resolution": "1k",
    "output_format": "png"
  }'

Edit with a reference image

curl -X POST 'https://sigrunai.com/v1/images' \
  -H 'Authorization: Bearer sk_xxx' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "gpt-image-2.6-sunburst",
    "prompt": "Keep the product, switch to a soft studio background",
    "quality": "high",
    "size": "16:9",
    "images": ["https://example.com/product.jpg"]
  }'

Success response:

{
  "code": 20000,
  "msg": "提交成功",
  "data": {
    "task_id": "img_xxxxxxxx",
    "status": "queued",
    "progress": 0,
    "model": "gpt-image-2.6-flare",
    "cost": 0.4,
    "result": {}
  }
}

Authorizations

Authorization

string

header

required

控制台创建的 API Key，格式：Bearer sk_xxx

Body

application/json

model

string

required

gpt-image-2 (plus low/medium/high), gpt-image-2.6-flare / gpt-image-2.6-sunburst, nano-banana-2 / nano-banana-2.1 / nano-banana-pro, seedream-5.5, or muse-image

Example:

"nano-banana-2"

prompt

string

GPT Image 2 / Seedream 5.5 / Muse Image max 4000; GPT Image 2.6 max 32000; Nano Banana max 10000. Nano Banana / Seedream prompt is optional when images are provided. Muse Image prompt is always required

Required string length: 1 - 32000

quality

enum<string>

default:auto

GPT Image 2: auto / low / medium / high, billed by tier. GPT Image 2.6 also accepts xhigh / max without changing the price

Available options:

auto,

low,

medium,

high,

xhigh,

max

integer

default:1

GPT Image 2 accepts 1–4. GPT Image 2.6, Nano Banana, Seedream 5.5, and Muse Image accept 1 only

Required range: 1 <= x <= 4

ratio

string

size

string

Alias of ratio. GPT Image 2.6 accepts auto or an aspect ratio, not a pixel size. Muse Image reduces a pixel size such as 1024x1024 to its ratio

resolution

string

GPT Image 2.6 accepts 1k only. Nano Banana: 0.5k/1k/2k/4k depending on the model

images

string<uri>[]

Reference image URLs. GPT Image 2 max 15; GPT Image 2.6 max 16; Nano Banana max 5; Seedream 5.5 max 8; Muse Image max 10. GPT Image 2.6 svg and gif cannot include reference images

Maximum array length: 16

output_format

enum<string>

GPT Image 2.6 only. jpg is treated as jpeg. For gif, omit size or send auto

Available options:

png,

jpeg,

webp,

svg,

gif

style

string

Muse Image only. Free-text style such as photorealistic, anime, or cinematic

Maximum string length: 1000

negative_prompt

string

Muse Image only. What should not appear, kept apart from prompt

Maximum string length: 1000

smart_queue

boolean

default:false

When true, a very low recent success rate keeps the task queued until recovery (up to about 2 hours)

Response

200 - application/json

提交成功

code

integer

Example:

20000

msg

string

data

object

Show child attributes

Last modified on November 6, 2026

curl --request POST \
  --url https://sigrunai.com/v1/images \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "gpt-image-2",
  "prompt": "Minimal product still life, white background, soft light",
  "quality": "medium",
  "n": 1,
  "ratio": "1:1"
}
'

{
  "code": 20000,
  "msg": "<string>",
  "data": {
    "task_id": "<string>",
    "status": "<string>",
    "progress": 123,
    "model": "<string>",
    "created_at": 123,
    "completed_at": 123,
    "cost": 123,
    "error_message": "<string>",
    "result": {
      "image_url": "<string>",
      "video_url": "<string>"
    }
  }
}