The second selectable GPT Image 2.5 product: the same tiers, shapes, background control and credit prices as Flare, with this guide focused on editing your own images.
Sunburst and Flare share one contract on this platform: the same quality tiers, shapes, background control and reference limit at the same credit price. Choosing between them is a matter of trying both, and we do not claim a quality difference between the two.
This guide explains editing with reference images. Use text mode to start from a written description without uploading images.
Editing mode requires at least one reference image and accepts up to sixteen. Quality tier, shape and, at 1K, background are selectable exactly as in text mode.
Start from one image or a set of up to sixteen and describe the change: replace an element, restyle the scene, or keep a subject while everything around it changes.
A 20000-character prompt limit leaves space for layout instructions, on-image copy and styling notes in a single request.
1K, 2K and 4K with automatic backgrounds, so the editing workflow reaches the same tiers as text generation.
Keep the product, change the setting, lighting or mood described in the prompt.
Supply several images of the same subject so a new scene still shows the same person or mascot.
Reuse an approved composition and vary the copy, colour scheme or background across a set.
Use a button on this page. It opens the right workspace with this model already selected, in the language you are reading now.
Write what should happen in the shot or image. Concrete subjects, framing and action give the model more to work with than short labels.
Choose length or quality tier, shape and sound where offered, then read the credit quote for those exact settings before you generate.
Generation uses account credits. Cost depends on the model, quality and output size or duration. Reference images, video or audio may add charges in some modes. Check the workspace quote for your selected inputs and settings before generating; the pricing page lists credit packages.
See credit packagesNot in its controls or its price on this platform. Both are selectable GPT Image 2.5 variants with the same tiers, shapes, background control and reference limit. We do not claim a quality difference.
Between one and sixteen references per task.
Yes, but that is text generation rather than editing. Reference-based mode requires at least one image.
1K if you need an explicit solid or transparent background, otherwise any tier. Automatic background keeps 2K and 4K available.
The same per-image rate as text generation at the tier you choose. The workspace shows the amount before you confirm.
Other models on this platform that do a similar job, or that pair well with this one.
One of the two selectable GPT Image 2.5 products on this platform: text and reference-based image generation with transparent-background support at the 1K tier.
Read the guideThe FLUX 3 family product available on this platform: text generation and reference editing at 1K, 2K or 4K, with PNG, JPEG and WebP output.
Read the guideNano Banana 2.1 covers both workflows on one model: images from a description, or edits driven by up to ten of your own reference images, at 1K, 2K or 4K.
Read the guide