Scenario/Problem:
In Canvas mode, when a user wants to generate an image based on similar parameters to an already generated image, the user can right-click an image and copy the text prompt. However, since references images are part of the prompt. the text is only half the battle. When copying the text prompt, the user will have to remember, find and restore the reference images.
(when I say ‘restore’ here, I mean re-fill the input form)
Suggestion:
Provide an easier way to ‘branch’ or ‘retry’ an image, for example one or more of:
Add a copy-paste flow or or one-click restore that includes both prompt and reference images
Add a way to restore prompt, reference images AND add generated image to reference images in a way that the model know which is which and you refine the text prompt before generating again
Maybe model, aspect ratio and resolution should also be restored
Please authenticate to join the conversation.
Gathering Interest
Feature Request
Get notified by email when there are changes.
Gathering Interest
Feature Request
Get notified by email when there are changes.