Creative AI
FAQ
Common questions about Creative AI features, inputs, credits, and usage.
Q. What can Creative AI do?
It supports text-to-image, image-to-image, text-to-video, image-to-video, character swap, and motion reference.
It also provides multi-image generation and sketch-based inputs.
Q. How does multi-image generation work? How many images can I upload?
Multi-image mode supports 1–12 reference images plus a prompt to generate a single image.
For best results, use clear, reasonably sized reference images.
Q. Which modes support sketch input?
Sketch input is available for sketch-to-art, pose transfer, and guess-and-complete workflows.
Saved sketches are uploaded as image inputs automatically.
Q. How many credits does a generation cost?
The default cost is 2 credits per generation. Please refer to each workflow card for the exact cost.
Q. Can I make results public?
You can choose public or private at generation time. Public creations appear in Explore.
Private content is visible only to you by default.
Q. What is Canvas for?
Canvas is designed for multi-step orchestration and asset reuse. Use @ mentions to pick workflows or assets.
It helps you build repeatable creative pipelines.
Q. What input formats are supported?
Common image, video, and audio formats are supported. Check input hints for exact limits.
Q. How can I improve output quality?
Be specific about subject, style, composition, and lighting in your prompt.
For multi-image generation, use references with consistent style and viewpoints.
Q. How should I handle copyright and compliance?
Make sure your uploaded materials are properly licensed, and public content should comply with local laws and platform rules.