Image model

GPT Image 2 Generator

Start with a written prompt or attach reference images, choose the shape of the result and generate it in the same studio as your videos.

Create with GPT Image 2

Open Workspace with this model or workflow selected. Adjust settings, add your media, and create in one place.

Start Creating

Use text-to-image or image-to-image

Text-to-image accepts a prompt without media. Image-to-image requires one to 16 JPEG, PNG or WebP reference images up to 30 MB; the source remains attached to the private task record.

Choose the shape of the final image

GPT Image 2 supports Auto plus square, portrait, landscape and panoramic aspect ratios. Each task creates one finished image.

Review the image cost before generation

Credits depend on the selected model, resolution, duration and references. The studio shows the total before generation.

Keep each result in private Creations history

See the prompt, credit cost and result status in one place. Image tasks appear beside video tasks, while your source image remains private.

When GPT Image 2 fits the brief

Use GPT Image 2 when the task is driven by a detailed written direction or a single source image that needs a deliberate transformation. It suits product concepts, editorial visuals and layout-aware compositions where the prompt should spell out both the desired change and the elements to preserve.

State the subject, environment, camera or viewpoint, lighting and visual treatment.
For source-image work, separate preservation instructions from requested edits.
Pick the target shape first when the image will be used in a banner, feed or vertical placement.