GPT Image 2.5
GPT Image 2.5 offers Flare for everyday content creation and Sunburst for detail-focused editing. Generate from text or use a reference photo with either variant. For reference-based edits with other models, explore the AI Image Editor.
Use the default Flare variant to draft a launch graphic, explore a product setting, or test a concept. Start with a clear brief and compare the directions you want to develop. Describe the audience, subject, and intended use so each attempt explores a purposeful variation rather than an unrelated image.
Choose Sunburst for product refinements and campaign visuals that need more attention to materials, lighting, and fine detail. This variant prioritizes detailed work and can take longer to process. For an existing photo, focus your instruction on the change you need and identify the visual elements that should remain recognizable.
Create product scenes, campaign graphics, and visual concepts from a prompt or reference image.
Try GPT Image 2.5
Use a reference photo to explore new backgrounds, lighting, and product settings. Describe the details you want to retain, such as the product shape, color, and label, then review them against the original. A simple brief might ask for a soft studio background while keeping the packaging and camera angle. Use the result as a visual draft for a listing or promotion.
Create a visual draft for an event poster, launch announcement, or promotional campaign. Specify the headline, composition, and style in your prompt, then check the generated wording and layout before publishing. Include the exact headline and date, and describe which element should attract attention first. Leave space for any finishing touches you plan to add later.
Turn ideas for characters, interiors, and illustrated scenes into visual concepts. Guide the viewpoint, lighting, and mood with your prompt, and choose a suitable aspect ratio and output resolution for your project. A wide frame can suit an environment study, while a portrait frame can suit a character concept. Keep the brief focused when comparing different visual directions. Explore more models and visual directions in the AI Image Generator.
Create or edit your image in three simple steps.
Describe the image you want to create. For an edit, upload a reference and explain both the change and the details you want to keep.
Select Flare or Sunburst, then choose your resolution and aspect ratio. Review the displayed credit requirement before generating. Start at a resolution appropriate for your draft, then use a larger output when the project calls for it.
Preview the generated image, download the result you want to keep, or adjust your prompt to explore another version.
Explore product visuals, campaign drafts, and illustration concepts with GPT Image 2.5.
Have you tried GPT Image 2.5? Share your experience with us through Contact Us.
What to know about variants, references, output size, and credits.
You can create images from text or edit an uploaded reference photo on AIEnhancer. Both Flare and Sunburst support these workflows.
Start with Flare for everyday content and visual exploration. Choose Sunburst when fine detail and editing control are the priority; it can take longer to process.
Yes. Enter a text prompt to create a new image. A reference photo is optional and is useful when you want to edit an existing subject or scene.
Choose 1K, 2K, or 4K output, with square, portrait, landscape, and wide aspect ratios. You can also select auto for the aspect ratio. The default settings are 1K and 1:1. Select the frame for your intended placement rather than relying on a heavy crop afterward.
Check the credit requirement shown for your selected variant and output settings before submitting. AIEnhancer uses the same account and credit system as its other image models.
You can ask the model to preserve specific details, but exact preservation is not guaranteed. Compare faces, lettering, colors, and product shapes with your source before using the output.