GPT Image 2.6 Sunburst API · Text to Image

Generate images or GIF animations from a text prompt with GPT Image 2.6 Sunburst. Choose a static-image ratio and resolution, or use GIF frame controls, then query the task or receive a terminal webhook.

Developer API pricing

Provider: SeeAPI

USD estimate: $0.0037 / credit · Estimated equivalent, not an additional charge.

Text to image

Image output: 7,800 credits / 1M tokens≈ $29.23 / 1M tokens
Cached image input: 520 credits / 1M tokens≈ $1.95 / 1M tokens
Image input: 2,080 credits / 1M tokens≈ $7.79 / 1M tokens
Cached text input: 325 credits / 1M tokens≈ $1.22 / 1M tokens
Text input: 1,300 credits / 1M tokens≈ $4.87 / 1M tokens
Text output: 2,600 credits / 1M tokens≈ $9.74 / 1M tokens

Credits are charged based on actual usage. View the final credit usage in your results.

Request

POST/v1/generations

AuthorizationSend your API key in the Authorization header using the Bearer scheme.Authorization: Bearer ****************

Header Params

stringrequired

Stable key that makes retries of the same request idempotent.

Use a new key for each new generation request.
Maximum length: 255 characters.

max length: 255

Example:: order-20261110-001

Body Params

application/json

stringrequired

Model ID for the generation model.

Use gpt-image-2-6-sunburst.

Allowed:gpt-image-2-6-sunburst

Default:: gpt-image-2-6-sunburst
Example:: gpt-image-2-6-sunburst

stringrequired

Endpoint capability to invoke.

Use text-to-image for this document.

Allowed:text-to-image

Default:: text-to-image
Example:: text-to-image

stringoptional

Optional public Provider ID that selects the implementation for this request.

Omit this field to use the default Provider, seeapi. Allowed values: seeapi, fal.

Allowed:seeapifal

Default:: seeapi
Example:: seeapi

objectrequired

Provider-specific generation parameters validated against the selected Provider's public input contract.

The available fields and their rules depend on provider.

Provider parameters

SeeAPIseeapiDefault Provider

stringrequired

Instruction describing the image or animation to create. Be explicit about subject, composition, style, lighting, and any text to render.

Required; 1–32,000 characters.

min length: 1max length: 32000

Example:: A cinematic city at night

stringoptional

Static-image aspect-ratio preset.

Do not use this parameter to resize GIF images. When output_format is gif, use the size parameter instead

Allowed:auto1:13:22:34:33:45:44:516:99:162:11:221:99:213:11:3

Default:: auto
Example:: 16:9

stringoptional

Static-image resolution tier.

The static pixel table applies to PNG, JPEG, WebP, and SVG; GIF uses its own frame size behavior.

Allowed:1K2K4K

Default:: 1K
Example:: 1K

stringoptional

Output file format.

Allowed: png, jpeg, webp, gif, svg. JPEG cannot use a transparent background. GIF uses size, frame_count, and frame_duration_ms.

Allowed:pngjpegwebpgifsvg

Default:: png
Example:: png

numberoptional

JPEG/WebP compression level.

Optional integer from 0 to 100. Send only for jpeg or webp; omit for PNG, GIF, or SVG.

min: 0max: 100

numberoptional

Number of frames in a GIF output.

integer 2–16. Default: 9.

min: 2max: 16

numberoptional

Duration of each GIF frame in milliseconds.

Only for GIF; 20–2,000 ms in 10 ms steps. Default: 200 ms.

min: 20max: 2000

stringoptional

Defines the pixel dimensions of each frame in the generated GIF. This parameter is used only for GIF output. Pass auto or a value in WIDTHxHEIGHT format; both dimensions must be divisible by 16. If omitted, the service uses 512x512. Do not use aspect-ratio values such as 1:1

The selected frame size also determines whether all frames can fit into the internal sprite sheet. The packed sheet must stay within 3840 pixels per side, 8,294,400 total pixels, and a maximum 3:1 long-to-short side ratio. Requests that exceed these limits return 400.
512x512: suitable for 2–16 frames
768x768: suitable for up to 12 frames
1024x1024: too large even for 2 frames

Example:: 512x512

stringoptional

Requested image quality level.

Allowed:lowmediumhighxhighmaxauto

Default:: medium
Example:: medium

stringoptional

Planning and generation complexity.

Allowed:lowmediumhigh

Default:: high
Example:: high

stringoptional

Background rendering mode.

Transparent is not available for JPEG or SVG.
Only available when resolution is 1K.

Allowed:autoopaquetransparent

Default:: auto
Example:: auto

stringoptional

Content moderation strength applied to the generation.

Allowed:autolow

Default:: low
Example:: auto

falfal

stringrequired

Text prompt that describes the image to generate.

Required; provide 1–20,000 characters.

min length: 1max length: 20000

Example:: A cinematic city at night

stringoptional

Preset aspect ratio.

Use either an aspect ratio preset or custom dimensions

Allowed:auto1:1_HD1:13:49:164:316:9

Default:: auto

stringoptional

Output resolution or quality preset for the Provider.

Allowed values: auto, low, medium, high. Default: high.

Allowed:autolowmediumhigh

Default:: high

stringoptional

Controls the requested reasoning effort.

Allowed values: low, medium, high, xhigh, max. Default: high.

Allowed:lowmediumhighxhighmax

Default:: high

stringoptional

Background rendering mode.

Allowed values: auto, transparent, opaque. Some values depend on output_format.
Transparent background is unavailable for JPEG.

Allowed:autotransparentopaque

Default:: auto

numberoptional

Maximum number of image assets requested from the Provider.

The Provider may return fewer assets. Range: 1–4. Default: 1.

min: 1max: 4

Default:: 1

stringoptional

Image file format for the generated output.

Transparent background is unavailable for JPEG.

Allowed:jpegpngwebpgif

Default:: png

numberoptional

Compression level for JPEG or WebP output.

Available only when output_format is jpeg or webp. Range: 0–100.

min: 0max: 100

numberoptional

Number of frames for GIF output.

Available only when output_format is gif. Range: 2–64.

min: 2max: 64

numberoptional

Duration of each GIF frame in milliseconds.

Available only when output_format is gif. Range: 20–2000 ms.

min: 20max: 2000

objectoptional

Custom output dimensions grouped.

Provide width and height together for a custom size. Custom dimensions are unavailable for GIF output.

numberrequired

Custom output width in pixels.

Required with height for a custom size. Range: 16–3840 px. Custom size must contain 655,360–8,294,400 pixels, with an aspect ratio between 1:3 and 3:1. Unavailable for GIF output.

min: 16max: 3840

numberrequired

Custom output height in pixels.

Required with width for a custom size. Range: 16–3840 px. Unavailable for GIF output.

min: 16max: 3840

string <uri>optional

Absolute HTTPS URL that receives the terminal generation webhook.

Optional. Maximum length: 2048 characters.

max length: 2048

Example:: https://client.example.com/webhooks/seeapi

Responses

202Generation accepted and queued for processing. This is the initial POST /v1/generations response, not the final task result.

Bodyapplication/json

The initial status is processing; use generation.id with GET /v1/generations/{task_id} to query the task.
The terminal task response may include generation.input, generation.result.assets, consumed_credits, and completed_at; those fields belong to the task-status response, not this HTTP 202 response.

objectrequired

Envelope containing the accepted generation task metadata and initial credit counters.

Use generation.id with the task status endpoint to follow progress.

stringrequired

Unique task identifier used to query the generation status.

Store this value and pass it to the task details endpoint.

Example:: task_xxx

stringrequired

Model ID used for this generation.

Example:: gpt-image-2-6-sunburst

stringrequired

Endpoint ID used for this generation.

Example:: text-to-image

stringrequired

Public Provider selected for this generation.

Allowed:seeapifal

seeapi: SeeAPI
fal: fal

Example:: seeapi

stringrequired

Current task status. A 202 response uses processing.

Allowed:processing

Example:: processing

numberrequired

Public credits reserved before provider-cost settlement. Values can include up to three decimal places.

Example:: 20

numberrequired

Public credits reported at response time. This is typically 0 while the task is processing; query the task after a terminal status for actual usage.

Example:: 0

numberrequired

Public credits refunded at response time, if any, for the accepted task.

Example:: 0

string <date-time>required

Creation timestamp in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

string <date-time>required

Timestamp of the latest task update in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

booleanrequired

Whether this response is an idempotent replay of an earlier creation request; this field is part of the creation response.

Example:: false

200Idempotent replay of an existing generation request.

Bodyapplication/json

Returned when the same Idempotency-Key is reused with the same request body.

objectrequired

Envelope containing the accepted generation task metadata and initial credit counters.

Use generation.id with the task status endpoint to follow progress.

stringrequired

Unique task identifier used to query the generation status.

Store this value and pass it to the task details endpoint.

Example:: task_xxx

stringrequired

Model ID used for this generation.

Example:: gpt-image-2-6-sunburst

stringrequired

Endpoint ID used for this generation.

Example:: text-to-image

stringrequired

Public Provider selected for this generation.

Allowed:seeapifal

seeapi: SeeAPI
fal: fal

Example:: seeapi

stringrequired

Current task status. A 202 response uses processing.

Allowed:processing

Example:: processing

numberrequired

Public credits reserved before provider-cost settlement. Values can include up to three decimal places.

Example:: 20

numberrequired

Public credits reported at response time. This is typically 0 while the task is processing; query the task after a terminal status for actual usage.

Example:: 0

numberrequired

Public credits refunded at response time, if any, for the accepted task.

Example:: 0

string <date-time>required

Creation timestamp in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

string <date-time>required

Timestamp of the latest task update in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

booleanrequired

Whether this response is an idempotent replay of an earlier creation request; this field is part of the creation response.

Example:: false

400

Invalid JSON, parameters, Schema, or API Policy

401

Invalid API Key or authentication failed

402

Insufficient credits

403

The current request is not allowed

404

Model, Endpoint, or task not found

409

Idempotency conflict or the same request is still being created

413

Request body is too large

429

Too many requests

503

Service temporarily unavailable

Query Task Status

After submitting a task, use the unified query endpoint to check progress and retrieve results:

Get Task DetailsLearn how to query task status and retrieve generation results

For production workloads, prefer callbacks to polling so you can process terminal events as soon as they arrive.

Note

Use a fresh Idempotency-Key for each new generation request; reuse it only when retrying the same request body.

Related Resources

Market Overview

Explore all available models

Credits & Billing

View credit balance, plans, and billing activity

Usage

Review API usage and credit consumption

Callbacks

Manage the secret used to verify generation webhook signatures. Callback URLs are supplied when you create each generation.

Webhook flow

Callback URL per generation.

Webhook signing key

Code Examples

Request

const response = await fetch("https://api.seeapi.com/v1/generations", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Idempotency-Key": "order-20261110-001",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
  "model": "gpt-image-2-6-sunburst",
  "endpoint": "text-to-image",
  "provider": "seeapi",
  "input": {
    "prompt": "A cinematic city at night",
    "aspect_ratio": "16:9",
    "resolution": "1K",
    "output_format": "png",
    "size": "512x512",
    "quality": "medium",
    "complexity": "high",
    "background": "auto",
    "moderation": "auto"
  },
  "callback_url": "https://client.example.com/webhooks/seeapi"
})
});

const result = await response.json();

ResponseHTTP 202

{
  "generation": {
    "id": "task_xxx",
    "model": "gpt-image-2-6-sunburst",
    "endpoint": "text-to-image",
    "provider": "seeapi",
    "status": "processing",
    "precharge_credit": 20,
    "consumed_credits": 0,
    "refunded_credits": 0,
    "created_at": "2026-08-18T10:00:00Z",
    "updated_at": "2026-08-18T10:00:00Z"
  },
  "idempotent": false
}

const response = await fetch("https://api.seeapi.com/v1/generations", { method: "POST", headers: { "Authorization": "Bearer YOUR_API_KEY", "Idempotency-Key": "order-20261110-001", "Content-Type": "application/json" }, body: JSON.stringify({ "model": "gpt-image-2-6-sunburst", "endpoint": "text-to-image", "provider": "seeapi", "input": { "prompt": "A cinematic city at night", "aspect_ratio": "16:9", "resolution": "1K", "output_format": "png", "size": "512x512", "quality": "medium", "complexity": "high", "background": "auto", "moderation": "auto" }, "callback_url": "https://client.example.com/webhooks/seeapi" }) }); const result = await response.json();

{ "generation": { "id": "task_xxx", "model": "gpt-image-2-6-sunburst", "endpoint": "text-to-image", "provider": "seeapi", "status": "processing", "precharge_credit": 20, "consumed_credits": 0, "refunded_credits": 0, "created_at": "2026-08-18T10:00:00Z", "updated_at": "2026-08-18T10:00:00Z" }, "idempotent": false }

Responses

202Generation accepted and queued for processing. This is the initial POST /v1/generations response, not the final task result.

Bodyapplication/json

The initial status is processing; use generation.id with GET /v1/generations/{task_id} to query the task.
The terminal task response may include generation.input, generation.result.assets, consumed_credits, and completed_at; those fields belong to the task-status response, not this HTTP 202 response.

objectrequired

Envelope containing the accepted generation task metadata and initial credit counters.

Use generation.id with the task status endpoint to follow progress.

stringrequired

Unique task identifier used to query the generation status.

Store this value and pass it to the task details endpoint.

Example:: task_xxx

stringrequired

Model ID used for this generation.

Example:: gpt-image-2-6-sunburst

stringrequired

Endpoint ID used for this generation.

Example:: text-to-image

stringrequired

Public Provider selected for this generation.

Allowed:seeapifal

seeapi: SeeAPI
fal: fal

Example:: seeapi

stringrequired

Current task status. A 202 response uses processing.

Allowed:processing

Example:: processing

numberrequired

Public credits reserved before provider-cost settlement. Values can include up to three decimal places.

Example:: 20

numberrequired

Public credits reported at response time. This is typically 0 while the task is processing; query the task after a terminal status for actual usage.

Example:: 0

numberrequired

Public credits refunded at response time, if any, for the accepted task.

Example:: 0

string <date-time>required

Creation timestamp in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

string <date-time>required

Timestamp of the latest task update in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

booleanrequired

Whether this response is an idempotent replay of an earlier creation request; this field is part of the creation response.

Example:: false

200Idempotent replay of an existing generation request.

Bodyapplication/json

Returned when the same Idempotency-Key is reused with the same request body.

objectrequired

Envelope containing the accepted generation task metadata and initial credit counters.

Use generation.id with the task status endpoint to follow progress.

stringrequired

Unique task identifier used to query the generation status.

Store this value and pass it to the task details endpoint.

Example:: task_xxx

stringrequired

Model ID used for this generation.

Example:: gpt-image-2-6-sunburst

stringrequired

Endpoint ID used for this generation.

Example:: text-to-image

stringrequired

Public Provider selected for this generation.

Allowed:seeapifal

seeapi: SeeAPI
fal: fal

Example:: seeapi

stringrequired

Current task status. A 202 response uses processing.

Allowed:processing

Example:: processing

numberrequired

Public credits reserved before provider-cost settlement. Values can include up to three decimal places.

Example:: 20

numberrequired

Public credits reported at response time. This is typically 0 while the task is processing; query the task after a terminal status for actual usage.

Example:: 0

numberrequired

Public credits refunded at response time, if any, for the accepted task.

Example:: 0

string <date-time>required

Creation timestamp in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

string <date-time>required

Timestamp of the latest task update in ISO 8601 format.

Example:: 2026-08-18T10:00:00Z

booleanrequired

Whether this response is an idempotent replay of an earlier creation request; this field is part of the creation response.

Example:: false

400

Invalid JSON, parameters, Schema, or API Policy

401

Invalid API Key or authentication failed

402

Insufficient credits

403

The current request is not allowed

404

Model, Endpoint, or task not found

409

Idempotency conflict or the same request is still being created

413

Request body is too large

429

Too many requests

503

Service temporarily unavailable