Experiences
Create character pictures with our AI image generator
Start our image workflow with the companion you want to show. Then explain what belongs in the frame. A conversation may establish the story, but a picture request needs its own visible details.
In this article
Work from a companion to a finished request
Confirm that you are using the still-image option before accepting its cost. This request produces a picture, not audio or a moving clip.
- Find the subject in the image tool's character selector, using a category when needed.
- Write a description of a single frame, specifying the relevant clothes, posture, surroundings and light.
- Inspect how many images the request will make and the token cost shown.
- Submit the request, examine the picture and revise the main mismatch before another attempt.
Give the picture a coherent composition
Put the character and main visual idea at the beginning of the request. Describe a moment you could see in a single frame, leaving out location changes, incompatible poses and viewpoints that cannot share the same composition.
Review the generated picture against your intention. Hair, clothing and other specifics can differ from the request or shift across attempts, so do not depend on perfect visual consistency.
Control picture requests from the conversation
Where chat images are supported, open Ask and find Image Generation Toggle. Turn it off when you do not want that mode's image suggestions or generations; you will need to enable it again to use them.
Use this as an image-mode choice, not a spending switch for the whole service. Voice features, video and existing media unlocks still need their own cost checks.
Keep a result without assuming photo-editing access
Use a collection or download option when one is available for a completed image. If saving seems to fail, look for the result in your account first so you do not pay to generate it again unnecessarily.
Our character-and-text workflow is not a general-purpose editor for existing photographs. Do not assume it includes uploads, reference-photo changes or photo edits. If you encounter a separate input option, check what that tool accepts and whether you have permission to provide the material.
Questions and answers
Where do I find the token charge for a picture?
Read the cost on the generation action you are about to confirm. The tool, quantity and account choices matter; a token pack's size does not establish one price for every image request.
How do I request an image while chatting?
In a conversation that supports pictures, open Ask and check that image generation is enabled. Inspect the token requirement before submitting a picture request.
Does generation access give me permission to submit someone's photograph?
Access alone does not authorize using another person's image. Do not assume photo uploads are supported, and only provide material through an available input when you have the necessary rights and consent under our content rules.
Is an existing-photo editor included in the character image tool?
Select a character and describe the scene to create a new picture. That process does not by itself establish support for editing an uploaded photograph. If a separate editor offers an upload field, check its supported inputs and the rights and consent needed for your material.