You can create a video from a text prompt or, when supported, from an image.
Text-to-Video
Open AI Video and choose Text-to-Video or a text-based template.
Describe the subject, action, setting, mood, and visual style you want.
Use prompt enhancement if it is available and helpful.
Select a supported model or option when the flow provides one.
Review the displayed coin or access requirement.
Start generation and wait for processing to finish.
Watch the result, then save or share it when those actions are available.
Image-to-Video
Choose an image-based template.
Upload an image from your device.
Add a prompt describing the motion, camera movement, or transformation you want.
Select the available model and options.
Review the generation requirement and start processing.
The exact input types, model choices, duration, and options depend on the current AI Video release. Some requests may take longer or may be rejected if the input is unsupported or unsafe.