Creating Images in ChatGPT

ChatGPT offers a "Create image" option that appears when users click the "+" menu button next to the prompt bar. After selecting this option, the prompt changes to "Describe an image," allowing users to type a description of the desired picture.

Using Styles

A Styles menu provides quick access to preset visual themes such as Coloring Book, Cyberpunk, or Anime. Choosing a style inserts a pre‑written prompt that guides the AI toward that aesthetic. Users can type their own description beneath the style prompt before generating the image.

Changing Styles After Generation

Once an image is created, users can select a new style from the same menu. The new style prompt replaces the previous one, and ChatGPT applies the change while keeping the original image context, producing a revised version in the chosen style.

Editing Images

ChatGPT includes an image editor accessible by clicking the generated picture. In full‑screen mode, a brush tool lets users highlight areas to modify and then type new instructions in the prompt bar. Undo and redo buttons are also available.

Mobile App Experience

On mobile devices, the image creation flow mirrors the desktop process, though the Styles dropdown is absent. Users must describe the desired style in text, and a size‑adjustment slider helps fine‑tune edits.

Working with Text and Existing Photos

The model handles text well, enabling the creation of graphics that include accurate lettering, such as party invitations. Users can also upload their own photos and request changes—like altering backgrounds, applying artistic styles, or adding imaginative elements.

Saving and Sharing

After finalizing an image, users can hover over it (desktop) or tap it (mobile) to reveal options for downloading, copying, or sharing to platforms like X, Reddit, and LinkedIn.

Usado: News Factory APP - descoberta e automação de notícias - ChatGPT para Empresas