JAKARTA - ChatGPT is reportedly preparing a new feature called Image Styles which will make it easier for users to create or convert images into various popular visual styles such as anime, cyberpunk, coloring book, to Studio Ghibli theme.

This feature is known from the results of the latest beta version of the APK teasedown ChatGPT, although until now it has not been officially announced by OpenAI.

The Image Styles feature will allow users to choose certain aesthetic styles that have been provided. ChatGPT will automatically add detailed instructions to your prompt, so that the image results will immediately match the desired style without having to bother looking for or designing a complicated prompt yourself.

This feature can be used well for:

Transforming existing images into certain styles (for example: from photo to anime).

Create a new image directly with the selected style from the start.

According to the APKEARdown, here are some available styles (styles) along with their descriptions:

How This Feature Works

When this feature is available, users can:

Upload the image to ChatGPT and select one of the desired Image Styles. ChatGPT will automatically add a style instruction to produce images in selected styles.

Using a text prompt if you want to create a new image from scratch, just add another directive if you want certain details on the image results.

This style instruction can also be modified, for example, changing captions, adding accents, or requesting specific nuances to get results that really fit the user's taste.

The Image Styles feature has not yet arrived publicly and there has been no official announcement from OpenAI.

The prediction of this feature is based on the results of the application code Inspection (APKEARdown), so that the final version of the feature may be different when it is officially released.

With the presence of Image Styles, the experience of making 'photo Ghibli', anime, or cartoons from any picture on ChatGPT is promised to be much easier and more beginner-friendly. You can produce cool images without having to be good at compiling AI promotions or technicals.


The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)