Documentation

ChatGPT Image 2.0 Generator creates and edits AI images online. It supports text-to-image, image-to-image, image editing, mask editing, and image variations for posters, product visuals, social media graphics, concept sketches, and design previews.

Features

  • Text to image: generate images from a prompt.
  • Image to image: upload reference images or provide image URL/file_id references.
  • Image editing: modify uploaded images with a text prompt.
  • Mask editing: paint editable areas on the canvas or upload a mask image.
  • Variations: upload a square PNG image and create similar alternatives.

Settings

  • Model: gpt-image-2 is used by default; other image models can be selected when supported by the API.
  • Count: controls the number of returned images.
  • Size: supports auto, square, landscape, and portrait sizes.
  • Quality: supports auto, high, medium, low, and related model options.
  • Format: supports png, jpeg, and webp; transparent backgrounds cannot use jpeg.
  • Streaming preview: shows partial images and final images when enabled.
  • Advanced JSON: adds supported API parameters and must be a JSON object.

Tips

  • Describe the subject, scene, style, composition, lighting, and output purpose clearly.
  • Do not mix uploaded images with URL/file_id references in one image-to-image or edit request.
  • Variations require a square PNG input image.
  • Use png or webp when transparent background is required.

Privacy and Limits

Image requests are sent to the server API and forwarded to the image model service. Do not upload sensitive IDs, private photos, unauthorized assets, or content that violates platform rules.