GPT Image 2.5
GPT Image 2.5 from OpenAI helps your ideas take shape with lifelike detail and edits that stay closer to your intent. Create from text or refine a reference, choosing Flare for quicker exploration or Sunburst for careful finishing. Start creating with GPT Image 2.5 on Shotyard today.
Keep Your Vision Intact with GPT Image 2.5
Recognize the Subject in Every Variation
Explore another setting or style while giving faces, shapes, and defining details a stronger connection to your reference.
Make a Change Without Starting Over
Describe the part you want to adjust and the details you want to keep. Focused instructions help each revision stay close to your original idea.
Match the Model to Your Next Step
Use Flare to explore ideas quickly. Switch to Sunburst when a detailed image or a delicate edit deserves more attention.
What You Can Create with GPT Image 2.5
- •Convincing Image Detail: Give portraits, environments, and illustrations a more believable sense of light, surface, and depth.
- •Reference Image Fidelity: Use uploaded images to guide a new composition while retaining recognizable subject details.
- •Selective Image Editing: Request a change to an outfit, object, or background and specify what should stay in place.
- •Text and Layout Direction: Bring wording, placement, and a consistent visual style together in posters and illustrated graphics.
- •Flexible Output Sizes: Choose from 12 aspect ratios and 1K, 2K, or 4K resolution to fit the space where your image will appear.
Carry a Familiar Face into a New Look
A reference gives GPT Image 2.5 a specific starting point for an edit. This published example shows a change of clothing while the child remains recognizable against the blue portrait background. Use the same approach when you want a different look with continuity in the subject.

Refine Individual Details with GPT Image 2.5
Small adjustments can be harder than creating a new picture. GPT Image 2.5 is designed to follow targeted changes while retaining more of the surrounding image. This image-editing sequence compares revisions to a full-body portrait; it is a demonstration of still-image edits. On Shotyard, upload the image you want to revise and describe the next change.
Give Text and Illustration a Shared Structure
Describe the visual order as well as the subject. GPT Image 2.5 can combine a repeated layout, distinct scenes, and text within one composition. The national-park stamp example shows how a common illustration style can connect a set of different places. Specify the wording and arrangement in your own brief, then check the result before sharing.

How to Use GPT Image 2.5 on Shotyard
Start with an idea or an image you want to change.
Describe the Image or Edit
Write your prompt and add reference images if needed. For an edit, explain what to change and what to preserve.
Choose Your Version and Size
Keep GPT Image 2.5 Flare selected or choose Sunburst from the model menu. Open the settings to set an aspect ratio and 1K, 2K, or 4K resolution.
Generate and Review
Check the displayed credit cost, generate your image, and review it in the workspace. To refine it further, use the result as a reference for another edit.
Choose the plan that's right for you
Get all features of Shotyard, complete tasks quickly and achieve professional results with our advanced AI technology.
Basic
- 300 credits/month
- All features available
- Unlimited downloads per day
- Assets owned by customer
- Faster generation speed
- Priority support
Pro
- 1,000 credits/month
- All features available
- Unlimited downloads per day
- Assets owned by customer
- Faster generation speed
- Priority support
Max
- 2,500 credits/month
- All features available
- Unlimited downloads per day
- Assets owned by customer
- Faster generation speed
- Priority support
Prime
- 7,000 credits/month
- Team management
- Up to 5 team members
- All features available
- Unlimited downloads per day
- Assets owned by customer
- Faster generation speed
- Priority support
GPT Image 2.5 Specifications
Available GPT Image 2.5 settings in the Shotyard image generator.
| Developer | OpenAI |
| Model versions | Flare and Sunburst; Flare selected by default |
| Generation modes | Text to image; image editing with optional references |
| Output resolution | 1K, 2K, 4K |
| Aspect ratios | 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, 21:9, 27:16, 16:27, 9:8, 8:9 |
| Reference images | Up to 16 images; up to 20MB each |
| Reference formats | JPEG, PNG, WebP |
| Prompt length | Up to 20,000 characters |
Last updated: September 9, 2026
Questions About GPT Image 2.5
What is GPT Image 2.5?
GPT Image 2.5 is OpenAI's image generation and editing family released in September 2026. It creates images from written instructions and can use existing images as references for changes.
Should I choose Flare or Sunburst?
Flare is suited to everyday creation and quicker exploration. Sunburst focuses on more demanding visual work and editing precision, with longer generation times. Both are available in the model selector on this page.
How does GPT Image 2.5 differ from GPT Image 2?
OpenAI describes improvements in reference fidelity, lighting, textures, and control over edits. Flare also emphasizes faster generation. The best choice depends on your prompt and the details you need to preserve.
Can I edit an existing image?
Yes. Add the image as a reference and describe your intended change. You can upload up to 16 JPEG, PNG, or WebP images, with a limit of 20MB per image.
Which resolutions and aspect ratios can I select?
Both versions offer 1K, 2K, and 4K resolution on Shotyard. The 12 aspect ratios are 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. Open the generator settings to choose them.
How is GPT Image 2.5 charged on Shotyard?
Generation uses your Shotyard credits. The generator displays the cost for your selected version and settings before you submit; check that amount when changing the resolution.
How can I get more accurate edits?
Identify the exact object or area to change, describe the desired result, and list the parts to keep. For another revision, use the latest result as your reference so the next request starts from that image.
What should I include in a detailed prompt?
State the subject, composition, lighting, and style. Add the exact wording and its position when the image includes text. The prompt field accepts up to 20,000 characters, so there is room for a structured brief.
Where can I use GPT Image 2.5?
You can use GPT Image 2.5 on Shotyard.
