Images API

AI image generationand editing API

Generate and edit images with GPT Image 2.5. Create product visuals, campaign assets, and illustrations from text or reference images.

OpenAI-compatible
1K / 2K / 4K
Generation & editing

Try Images Playground Read the API docs

Cobalt-blue ceramic cup on a plinth in a softly lit pale-blue studio — PROMPT
A cobalt-blue ceramic cup, soft studio lighting, a clean pale-blue backdrop.

IMAGE CREATION

Text to image, reference edits, and flexible output.

Text-to-image generation

Describe the subject, scene, and lighting to create product visuals, illustrations, and campaign assets.

Reference image editing

Edit an existing image, change its setting or style, or use a mask to select an area.

Resolution and format

Choose the ratio, 1K / 2K / 4K, and file format. Up to 4 outputs per request.

IMAGE EDITING

Put your product in a different setting.

“Keep the blue cup. Move it to a sunlit terrace with warm tones and olive-leaf shadows.”

Blue ceramic cup in a pale-blue studio — Reference image

The same cup on a warm stone surface with leafy shadows — Edited scene

Image editing exampleTry your own reference

FROM BRIEF TO IMAGE

Create images around the work you need to publish

Generate a new concept or edit an existing image

Text-to-image generation suits a new illustration, campaign concept, or article cover when there is no source image to preserve. Describe the subject, setting, composition, and visual style so the output has a clear purpose.

Use reference image editing when you already have a starting point. Specify what should stay and what should change, such as keeping a product while replacing its setting. Review the result for product details, proportions, and brand accuracy before publishing.

Explore visual content workflows

Plan the composition for its destination

A website banner, social post, and product card need different framing. Choose the aspect ratio for the intended placement, describe where the subject should sit, and leave room for any headline or interface elements you will add later.

Use resolution to match the final output size, not to compensate for an unclear brief. For a series of assets, reuse the same descriptions of lighting, palette, and composition while changing the subject or scene. Check the series together for visual consistency.

Browse image prompt examples

Test a creative direction before automating it

Start in the browser to compare prompts, reference images, and output settings. Once a direction works for your content, use the API to bring the same inputs into your application or publishing workflow.

Automation can prepare candidate images for review, but generation does not make every output publication-ready. Keep a review step for visible text, logos, and product features, especially when an image will represent a real item for sale.

Try the AI image generator

INTEGRATION

Image generation for your workflow.

API
API docs
Generate and edit images in your app or automated workflow.
Playground
Open Playground
Try prompts, reference images, and output settings in your browser.
MCP
MCP guide
Give compatible AI agents access to image generation and editing tools.

Create a taskcURL

curl --request POST 'https://api.sukidata.com/v1/images/generations/async' \
  --header "Authorization: Bearer $SUKIDATA_API_KEY" \
  --header "Idempotency-Key: $SUKIDATA_REQUEST_ID" \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt-image-2.5",
    "prompt": "A cobalt-blue ceramic cup, soft studio lighting",
    "size": "1254x1254",
    "quality": "high",
    "n": 1
  }'

Retrieve a taskcURL

curl 'https://api.sukidata.com/v1/images/tasks/YOUR_TASK_ID' \
  --header "Authorization: Bearer $SUKIDATA_API_KEY"

AI IMAGES

AI image generation and editing prices.

GPT Image 2.5 · GPT Image 2

Manage AI balance

1K$0.03 / image

2K$0.05 / image

4K$0.08 / image

Priced by resolution. All quality settings cost the same.

Explore image ideas and prompts See pricing for all products

FAQ

Images API questions.

Can I edit an existing image?

Yes. Upload reference images in the Playground or send them to the image editing API. Describe the change you want; you can also supply a mask for a specific area.

Which models and settings are available?

GPT Image 2 and GPT Image 2.5 are supported. Both models support up to 4 outputs per request, auto, low, medium, or high quality, and PNG, JPEG, or WebP output. Choose an aspect ratio and resolution in the Playground, or set dimensions in the API.

How is image generation billed?

Pay per image from your USD AI balance: $0.03 at 1K, $0.05 at 2K, or $0.08 at 4K.

How do I wait for an image?

Submit an asynchronous request to receive a task ID, then check the task until it completes or fails. The Playground follows the task for you; MCP provides tools for starting a task and checking its result.

START CREATING

Create images for your products and content.

Generate or edit images in the Playground, then bring them into your product through the API.

Open Playground Read the docs