Available nowFeatured

GPT Image 2.5 Sunburst: image editing

By OpenAI

OpenAI's image model for detailed briefs and careful edits. Generate from text, or edit up to 16 images with an optional mask, with a transparent background when you need one.

Model ID
Maker
OpenAI

Input

Fields marked * are required. Review the price before generating.

Example prompt. Edit it or write your own.

The images to use as a reference for the generation. A maximum of 16 images are allowed.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB · 0/16

White marks regions to edit and black marks regions to keep.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

Use 'auto' to infer from input images.

Additional settings

Higher settings increase detail, latency, and token usage. Use 'auto' to let the model choose.

Compression level from 0 to 100.

Starting price$0.10 / image

You sign up first. Your settings are kept.

About this model

Overview

GPT Image 2.5 Sunburst turns a written brief into up to 10 images per run, in square, portrait or landscape presets or a size of your own. Quality runs from low to max, and the background can be transparent for cutouts and stickers.

The edit mode takes up to 16 images. Add a mask to keep the change inside one area: white marks what to edit and black what to keep.

What you can build

  • Product images on a transparent background for online shops
  • Posters and covers with a short headline
  • Food and editorial photos from a written brief
  • Local fixes in a photo, limited to the masked area
  • Several rounds of edits on one image

Prompting tips

  • Put any words that should appear in the image in quotes and say where they go.
  • Try ideas at low quality, then render the final image at high or max.
  • With a mask, describe only what should appear in the white area.

What to check

Check fine texture, repeated objects and any requested text. For an edit, specify one change at a time so you can judge which instruction made the difference.

Inputs

  • Prompt (required): The prompt for image generation.
  • Images (required): The URLs of the images to use as a reference for the generation. A maximum of 16 images are allowed.
  • Mask (optional): Mask image URL. White marks regions to edit and black marks regions to keep.
  • Image Size (optional): The size of the generated image. Use 'auto' to infer from input images. Options: Square HD, Square, Portrait 4:3, Portrait 16:9, Landscape 4:3, Landscape 16:9, Auto (size of your image).
  • Background (optional): Background for the generated image. Options: auto, transparent, opaque.
  • Quality (optional): Quality for the generated image. Higher settings increase detail, latency, and token usage. Use 'auto' to let the model choose. Options: auto, low, medium, high, xhigh, max.
  • Number of Images (optional): Number of images to generate. From 1 to 10.
  • Output Format (optional): Output format for the images. Options: jpeg, png, webp.
  • Output Compression (optional): Compression level from 0 to 100.

Output and limits

  • Image sizes: Square HD (1024×1024), Square (512×512), Portrait 4:3 (768×1024), Portrait 16:9 (576×1024), Landscape 4:3 (1024×768), Landscape 16:9 (1024×576), Auto (size of your image).
  • File formats: JPEG, PNG, WEBP.
  • Up to 10 images per run.
  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • auto · up to / max · up to: $0.72 / image
  • high · up to: $0.24 / image
  • low · up to: $0.10 / image
  • medium · up to: $0.13 / image
  • xhigh · up to: $0.37 / image