GPT Image 2.5 Faster Generation, Precise Edits
Generate and edit images with GPT Image 2.5. Use Flare for fast everyday work, or Sunburst when you need tighter control across detailed edits.
Speed when you want it, precision when you need it
GPT Image 2.5 splits into two modes: Flare for fast iteration, and Sunburst for high-precision generation and editing.
Flare for fast iteration
Flare is the default for most work. It generates faster than GPT Image 2, so you can explore layouts, styles, and variations without waiting.
Sunburst for precise edits
Sunburst is built for tighter control. Change one detail in a complex image while keeping the rest of the scene intact.
Image-to-image editing
Upload a reference image and describe the change. GPT Image 2.5 follows the instruction while preserving identity, lighting, and composition.
1K to 4K output
Choose 1K, 2K, or 4K, plus extra aspect ratios and transparent, opaque, or auto backgrounds for production-ready assets.
Where GPT Image 2.5 shines
From product shots to campaign edits, GPT Image 2.5 covers fast drafts and high-precision finishing in one workflow.
Photoreal portraits and scenes
Generate lifelike portraits, product shots, and realistic scenes with natural lighting and recognizable identity.
Illustrations and concept art
Create stylized illustrations, posters, and editorial visuals with stronger text and layout control.
Targeted image editing
Replace a background, restyle clothing, or change the weather while keeping the subject, camera angle, and room intact.
Branded assets and mockups
Produce on-brand marketing visuals, packaging concepts, and campaign stills that stay consistent across variations.
Social and content creation
Make posts, banners, and thumbnails in the exact ratio you need, including transparent backgrounds for design overlays.
From prompt to polished image
Start from a prompt or a reference image, pick Flare or Sunburst, then generate at the resolution and ratio you need.
Upload an image or start fresh
Upload a reference image to edit, or start from scratch with a text prompt. Multiple references are supported.
Choose Flare or Sunburst
Use Flare for speed, or switch to Sunburst when the edit needs tighter control. Set ratio, resolution, and background next.
Generate and download
Generate one or more images, review the result, and download it. Failed jobs automatically refund your credits.
Common Questions & Answers
GPT Image 2.5 is OpenAI's latest image model on Nano Banana. It supports text-to-image and image-to-image, with two quality modes: Flare for speed and Sunburst for precise edits.
Flare is the default for most work: faster generation and everyday creative iteration. Sunburst takes longer and is better when you need tighter control, like changing one detail without shifting the rest of the image.
Yes. Upload one or more reference images and describe the change. GPT Image 2.5 can restyle, replace a background, or adjust details while keeping the parts you want unchanged.
GPT Image 2.5 supports auto, 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, 21:9, 27:16, 16:27, 9:8, and 8:9. Resolution can be 1K, 2K, or 4K. The 27:16, 16:27, 9:8, and 8:9 ratios are 1K only.
Credits depend on resolution: 1K = 4 credits, 2K = 6 credits, 4K = 12 credits. Flare and Sunburst use the same credit cost. Generating multiple images multiplies the total.
Choose Auto to let the model decide, Opaque for a solid background, or Transparent when you need a cutout-style PNG for design work.
GPT Image 2.5 requires credits. You can try it with your daily free credits after login, or top up with a plan or one-time credit pack.