OpenAI dropped a newer version of ChatGPT Images, now powered by GPT Image 1.5. It's showing up in ChatGPT and available through the API.
Faster results. It makes pictures up to 4x quicker now. More accurate edits. It sticks to what you ask, only changes what you say, and keeps things like lighting and faces looking the same. Keeps the small stuff. Details like faces and layout stay solid even after edits. Creative tools. You can change a photo into a poster, ad, or artsy version using styles or short prompts.
Here’s what you can do. Edit photos you upload by adding stuff, removing things, or mixing parts together. Make new images in different styles from nothing. Use the presets and filters in the new Images tab to mess around with ideas without writing much.
For developers and teams. Same new features work with the GPT Image 1.5 API. Image use costs around 20% less than before, so you get more tries for less money. Business use is rolling out after the general release.
If you'd like to access this model, you can explore the following possibilities:
