Edit photos with GPT Image 2 by just describing the change, free to tryTry It Free

Reference images (optional)

For the best result
  • Optional: upload an image to edit, plus up to 3 more as references. With no image, your prompt alone makes the picture
  • PNG, JPG or WEBP, up to 10 MB each
  • With a photo, say what to change and what to keep
0/5000
Aspect ratio

GPT Image 1

GPT Image 1 is retired at OpenAI and not live here yet. Leave your email, or edit a photo with a newer model: one free image, no sign-up.

Your image will appear here.

GPT Image 1 status, in plain facts

OpenAI has shut it down

OpenAI's model page says it was shut down on October 23, 2026 and is no longer available. It points to GPT Image 2.5 for current image generation and editing.

Source

This site can't run it yet

The tool above shows the full form, but pressing the button opens a coming soon dialog instead of making an image. This site is independent and not operated or endorsed by OpenAI.

The waitlist costs nothing

The dialog asks for an email and what you plan to make. Nothing is uploaded or charged, and it does not use your free image.

Your free image works on newer pages

One free image per visitor, no sign-up and no card, counts across the whole site. Spend it on the 1.5 or 2 page, the closest models that run here.

GPT Image 1 coming soon placeholder: not available on this site yet

What GPT Image 1 was

It began as GPT-4o image generation, shown on March 25, 2025, and reached the API as GPT Image 1 on April 23, 2025. It was the first model of the GPT Image series and the successor to DALL-E.

It took text and images as input, so it could edit a photo as well as create one.

What changed after the first model

GPT Image 1.5 followed in December 2025. OpenAI described it as improving edit precision, instruction following and text rendering compared with GPT Image 1.

GPT Image 2 added reasoning before it draws, and 2.5 came in September 2026. If you searched this name to edit a photo, a newer page does the same job today. Upload the photo, say what to change and what to keep, and check the result against your original.

Both newer pages work the same way. Drop in a JPG, PNG, WEBP or iPhone HEIC photo, up to 10 MB each; the browser scales it down before upload. The 1.5 page takes the photo plus up to 3 references and returns a 1:1, 3:2 or 2:3 image. The 2 page takes up to 15 references, such as a product, a face or a style to copy, and offers Auto plus seven ratios, up to 4K. Write the change in the prompt box, up to 5,000 characters, then download the result: there is no history to come back to.

GPT Image 1 vs 1.5 vs 2: which one to use

Pick by what runs today, not by the name you remember.

First modelGPT Image 1.5GPT Image 2
At OpenAIShut down Oct 23, 2026Successor to the first modelAdded reasoning, April 2026
On this siteComing soon, waitlist onlyRuns, first image freeRuns, first image free
Photos per run hereUp to 4 once liveUp to 4Up to 16
Output here1:1, 3:2 or 2:3, once live1:1, 3:2 or 2:3Auto or 7 ratios, at 1K
Best forNothing: it is retiredEdits with up to 3 referencesEdits that need more reasoning

Frequently asked questions

It is OpenAI's first GPT Image model, the one behind GPT-4o image generation. It creates images from text and edits images you give it. It reached the API as GPT Image 1 on April 23, 2025.

Not yet. The page shows the tool, but this site can't run the model, so the button opens a coming soon dialog and no image is made. For an edit today, use the 1.5 or 2 page.

No. OpenAI's docs say it was shut down on October 23, 2026 and recommend GPT Image 2.5 instead. That is why it is not live here either.

The 1.5 model came first, then GPT Image 2 and 2.5. This site runs 1.5 and 2 on their own pages, plus a fast version of 2.5. Each page runs one model, so the page you open decides what you get.

No. DALL-E 3 was OpenAI's earlier image model, and GPT Image 1 succeeded it. Neither can be used on this site today.

No. GPT Image 1 Mini is a separate model released on October 6, 2025. This page is about the full model, and this site doesn't run Mini either.

No to both. Joining only stores your email and your answer to what you'd like to make. The free image stays available for the 1.5 or 2 page.

GPT Image 1 is not available on this site. What is free here is one image on a working page, with no sign-up and no card. Your photo is only used to create your images and is never used to train AI models.

Open the 1.5 page if you want the closest successor, or the 2 page if the edit needs more reasoning and more reference photos. Both take a photo plus a prompt that says what to change and what to keep.

It answers what happened to GPT Image 1 and where to edit a photo today: the GPT Image 1.5 or GPT Image 2 page.