Skip to main content
TikTomato
Sign in

GPT Image 2.5 image generator

Run OpenAI's GPT Image 2.5 (Flare) from your browser. Write a prompt, or add up to 4 reference images and describe the change. Free credits when you sign up, no API key, nothing to install.

Reference images (optional) · Up to 4 images, JPG / PNG / WebP, 5 MB each

GPT Image 2.5 vs GPT Image 2: what we measured

We ran both models ourselves rather than repeating the release notes. Ten calls, same prompts, same parameters, billing read back from the provider's own usage record.

GPT Image 2GPT Image 2.5
Provider cost, 1K$0.015$0.015
Output size, 3:4 at 1K1086 x 14481086 x 1448
Generation time, median of 571s68s
Two-line sign text, spelled correctly5 of 55 of 5
Pixels at 2K, same prompt6.29 MP4.18 MP

At 1K the two models came out even on every measure we could put a number on, and we could not pick a winner by eye either across five pairs of images. The one clear difference sits at 2K, where GPT Image 2.5 costs twice as much and returns fewer pixels for the same prompt. That is why this page runs at 1K, and why our main image generator still uses GPT Image 2 by default.

Timings are provider-side medians over five runs per model. Single runs are misleading here: one 1K image took anywhere from 60 to 116 seconds, a spread wider than the gap between the two models.

Made with GPT Image 2.5 on this page

Both images below came out of the model this page uses, at 1K, with the prompts shown underneath.

Enamel shop sign reading TIKTOMATO above a European street, generated by GPT Image 2.5
“A vintage enamel shop sign hanging above a narrow European street. The sign reads "TIKTOMATO" on the first line and "PHOTO RESTORATION - EST. 1998" on the second line, in clean serif lettering. Warm late-afternoon sunlight, shallow depth of field, photorealistic.”
Photorealistic portrait of an elderly fisherman in a knitted sweater, generated by GPT Image 2.5
“Close-up portrait of an elderly fisherman with a weathered face and white stubble, wearing a hand-knitted wool sweater, soft overcast daylight, sharp focus on the eyes, natural skin texture with visible pores and fine wrinkles, shot on an 85mm lens.”

GPT Image 2.5 questions

What is GPT Image 2.5?

GPT Image 2.5 is OpenAI's image generation and editing model, released after GPT Image 2. OpenAI describes it as sharper, with stronger reference fidelity and more natural lighting and textures. It comes in two variants, Flare and Sunburst; this page uses Flare.

Do I need an OpenAI account or an API key?

No. Sign in with Google or Apple and you get free credits to start. Nothing to install and no API key to manage.

Is GPT Image 2.5 better than GPT Image 2?

In our own tests at 1K, no: same price, same output size, same speed, and we could not tell the images apart. The measured numbers are on this page. We publish them because most write-ups just repeat the release notes.

What resolution do I get?

1K, which for a 3:4 image is 1086x1448 pixels. We limit this page to 1K because that is the tier where GPT Image 2.5 costs the same as GPT Image 2. At 2K and 4K it costs twice as much and returns fewer pixels, so we do not offer it there.

Can I edit an existing image?

Yes. Add up to 4 reference images and describe the change. With references the request goes to the image-to-image variant of the same model.

How many credits does one image cost?

The same as a 1K image anywhere else on the site. The exact number is shown next to the generate button before you submit.