How to use ImageGPT v3.5

Edit an image using up to seven references.

References

Upload between one and seven images and refer to them in upload order as <image1>, <image2>, ... <image7>. The first image is the main image and determines the output proportions. The remaining images can define an identity, outfit, pose, background, object, palette, or style.

Writing a prompt

A strong request describes: what to change → where to change it → which reference supplies the details → what must stay unchanged → how lighting and style should match.

Assign an explicit role to every reference.
Name the exact subject and image number instead of using “he,” “she,” or “it.”
List the features that must not change, such as the face, pose, composition, clothing, or background.
Put exact poster or sign text in double quotes and specify its color, typeface, and perspective.
Split complex edits into stages: handle the character and outfit first, then the background and lighting.

Several panels in one image

The model can build a character sheet, a grid of expressions or poses, or a storyboard in a single image. To keep the result coherent:

State the layout: the number of panels, the grid (for example, 3×3), and the panel order from left to right and top to bottom.
Describe the content of each panel separately.
Require the same character appearance in every panel.
End by saying that the result is a single image containing all panels.

Examples

Preserve the face, pose, and body proportions of the person in <image1>. Replace only the outfit
with the outfit from <image2>, accurately transferring its material and details while matching the
lighting in <image1>.

Preserve the subject from <image1>. Replace the background with the scene from <image2> and apply
its lighting naturally to the face, hair, and clothing. Keep everything else unchanged.

In <image1>, replace the sign text with "OPEN 24/7". Preserve the original color, letter weight,
angle, perspective, and glow. Keep the surrounding architecture unchanged.

Create a four-view character sheet of the character in <image1>, from left to right: a close-up
facial portrait, a full-body front view, a side view, and a back view. Use a neutral gray
background. Keep the appearance, outfit, and proportions identical in every view.

Create a 3×3 grid of stickers of the person in <image1>. Expressions from left to right and top to
bottom: happy, sticking out tongue, furious, sad, annoyed, frightened, crying, smiling, shy. Keep the
person recognizable in all nine panels. The result must be a single image with a 3×3 grid.

Make the character from <image1> hold and present the product from <image2>. Professional studio
product photography with realistic lighting and soft shadows. The product is clearly visible and
sits naturally in the hands; lighting and colors match on the character and the product.

Replace the head and hairstyle of the character in <image1> with those of the character in <image2>.
Match the skin tone, head size, proportions, and lighting to <image1>. Keep the body, pose,
clothing, and composition of <image1> unchanged.

Fully restore and colorize the old photograph in <image1>. Remove scratches, creases, stains, and
noise, and rebuild sharp facial features, hair, and fabric texture. Apply natural colors and
realistic skin tones. Strictly preserve the face shape, features, expression, pose, clothing,
framing, and camera angle so the person stays recognizable.

Mode

High is the default mode, with the best quality and detail.
Economy is faster and cheaper. Use it for drafts and prompt tuning, then create the final version in High mode.

Resolution

Choose the output resolution: 1K is about 1.6 megapixels (1280×1280 for a square), 2K is about 4 megapixels (2048×2048 for a square). The result keeps the proportions of the first image, so there is no aspect ratio to choose. References are automatically resized to the selected resolution while preserving their proportions. A single run can produce up to two results.