Experiences
Create character images with Candy AI
Our image tools turn a visual request into a still picture. Choose or create the character first, then describe the scene you want rather than relying on a chat message to imply every visual detail.
Choose the subject before writing the scene
A still-image request does not create a voice call or video. Keep the intended format clear before you confirm a charge.
- Open the image-generation character picker and choose the appropriate character or category.
- Describe one scene, including the outfit, pose, setting and lighting that matter to you.
- Check the quantity and token charge displayed by the current tool.
- Generate, review the result and change the detail that needs correction before trying again.
Make the request visual
A concise scene description is easier to refine than a long narrative with several locations and actions. Put the main subject first and avoid asking for conflicting poses or multiple camera positions in one picture.
Exact clothing, hair and other details can vary between results. Treat a first image as something to review, not as a guarantee that every word has been reproduced precisely.
Use image requests in chat intentionally
Supported chats can also offer picture requests. The Ask menu has an Image Generation Toggle. Switching that mode off removes its image suggestions and prevents image generation through that mode until you turn it back on.
That switch is not a global setting that removes charges from every media feature. Review voice, video and prepared-content costs independently.
Keep the result and the request together
After a successful generation, use the available collection or download control to keep the image you want. If a result fails to save, check whether it appears in the account before requesting another charged generation.
Creating a character picture from text is different from uploading or editing an existing photo. Choosing a character does not automatically include a separate photo editor. Use an upload or editing option only when the selected tool offers it, and provide only material you have the right to use.
Questions and answers
How many tokens does one Candy AI image cost?
Use the amount shown by the selected generation action. Quantity, tool and account options can affect a request; do not infer a universal image price from the size of a token pack.
Can I generate pictures from the conversation?
Supported character chats offer image requests. Check the Ask menu and its image-generation setting, then review any token charge before sending the request.
Does the image generator let me upload any real person's photo?
No general permission follows from access to generation. Use only a supported upload control and material you have the right to provide, and follow our consent and content restrictions.
Can I upload a photo or edit an existing picture?
The character-image workflow starts with a chosen character and a written prompt. We do not promise a general-purpose photo-upload editor with that access. If a separate tool offers reference images or edits, check its supported inputs and permissions before using it.