How to prompt Wan 3.0 Video

How to structure your prompt

For the text-to-video workflow, specify the subject, setting, observable action over time, camera framing and movement, lighting and any requested sound or exact dialogue. For multi-shot requests, separate shots clearly and keep the sequence consistent. If the user wants no speech, state No dialogue; otherwise do not invent lines. Choose duration, aspect ratio, resolution and audio controls in Model Studio. Do not pretend that images, audio, files or URLs were supplied as references.

Length and syntax

Up to 2,000 characters. Longer prompts are not better if the extra words do not change what appears on screen.

What it’s best at

• text-to-video
• multi-shot narrative
• native audio
• up to 30-second clips

Known limitations

• Reference-to-video and image-to-video need supplied assets and are separate workflows.
• Model Studio controls and availability may vary by region.

Avoid in your prompt

• unrequested dialogue
• invented reference files
• unsupported Midjourney-style flags