GPT Image-2 Generator
Start with one headline and three short callouts so the first run proves text hierarchy instead of dense layout ambition.
Check: Useful when the first result must prove text hierarchy and spelling.
Open poster prompt
Use an existing packshot as the source of truth, then add a campaign layout around it without changing the product.
Check: Check labels and any generated spec text before using the result.
Open product edit
Create a product-first campaign image with one headline and short benefit labels that stay readable.
Check: Use when a product ad needs short approved copy before scaling variants.
Open campaign prompt
Turn one product into a reviewable contact sheet for hero, detail, scale, specs, and lifestyle slots.
Check: Best for ecommerce teams that need a full image set before SKU upload.
Open PDP set prompt
Use the starter-pack trend in a brand-safe way: fictional persona, readable pack title, no real likeness.
Check: Good viral hook when the concept is original and not based on a real person.
Open starter pack
Keep product variants aligned in scale, crop, shadow, and label treatment before uploading store media.
Check: Useful when SKU variants need consistent framing before store upload.
Open variant prompt
Create a readable product-screen concept for landing pages, sales decks, or feature reviews.
Check: Useful when a team needs a visual product concept before design polish.
Open UI prompt
Draft a marketplace content block with approved copy and clear benefit hierarchy.
Check: Use when copy is approved but the module layout still needs a draft.
Open A+ prompt
Create a phone-style photo with factual labels and arrows without turning it into a dense infographic.
Check: Good when the edit layer is the feature, not a full scene rewrite.
Open annotation prompt
Improve a profile portrait while preserving identity, crop intent, and natural expression.
Check: Add only when the source portrait is rights-cleared and identity must stay stable.
Open headshot edit
Generate a nostalgic multi-frame product lifestyle post with short readable frame labels.
Check: Good social format, but keep brands and real people out of the prompt.
Open collage prompt
Create a multi-view sheet before production so identity drift is visible before a batch of scenes.
Check: Use before batch scenes to catch identity drift early.
Open reference sheet
Use GPT Image-2 for the static ad layout first, then animate only after product and copy pass review.
Check: Use when copy and layout need approval before video.
Open ad prompt
Turn verified facts into a visual brief without letting the image invent numbers.
Check: Best when every number must map back to an approved source.
Open infographic prompt
Make a shareable process graphic with a short title, four steps, and mobile-readable labels.
Check: Useful for viral explainers where text readability is the point.
Open infographic
Plan product angles and ad beats as still frames before making a video.
Check: Use when video should start from approved still-frame direction.
Open storyboard prompt
Plan product ad beats as still frames so the product, camera rhythm, and labels are approved before video.
Check: Good for reference-first product video planning and ad beat review.
Open storyboard prompt
Use an inside-environment angle to make a product feel native to the scene while keeping the headline short.
Check: Matches inside-environment product ad workflows.
Open fridge POV prompt
Use a small line-art layer to make a product photo feel social without hiding the product.
Check: Good for social product images where the product still needs to stay clear.
Open doodle prompt
Build a product poster around one hero pack, local cues, and a clear flavor path.
Check: Useful when a poster needs product hierarchy before decorative world-building.
Open snack poster
Create a premium beauty product still while keeping label, material, and claims under control.
Check: Use for beauty concepts where material and label review matter.
Open skincare prompt
Use GPT Image-2 to turn a simple portrait idea into a softer illustration direction.
Check: Use when style direction matters more than product text accuracy.
Open style promptBefore generating
Yes, it is the main AuraTuner image model for poster-style layouts and readable copy, but every output still needs proofreading before use.
Yes. This page starts with text-to-image for a new visual. Switch to image-to-image in Studio when an existing product photo, poster, or layout should guide the result.
The starter setup uses a 1K image generation, 10 credits as of May 26, 2026 in AuraTuner.
AuraTuner supports 1K, 2K, and 4K for GPT Image-2. Auto aspect ratio is limited to 1K, and 1:1 does not support 4K.
Use text-to-image when you need a new campaign visual. Use image-to-image when product shape, packaging, or approved source photos need to stay close.
Avoid dense legal copy, unverified performance claims, or tiny comparison tables on the first run. Use a simpler proof first, then revise.