Text to video
Build a complete shot from a written brief—subject, location, action, lighting, lens, and mood.
Turn a prompt, a first frame, or a reference video into polished motion with Wan 3.0 Video.
Avoid copyrighted characters, logos, music, and sensitive, explicit, or violent content, or the provider may block the request.
Use the same prompt and settings to generate up to 4 video variants at once.
Wan 3.0 Video can turn a prompt, a first frame, or a reference video into 480P–1080P motion with optional audio.
One model, three creative paths
Wan 3.0 Video reads language, images, and motion references together. Direct the scene in natural language, preserve the visual cues that matter, and render social clips or polished campaign footage from one workspace.
Describe the subject, movement, camera, atmosphere, and sound. Wan 3.0 Video combines those signals into a coherent sequence with controllable framing and output quality.
Build a complete shot from a written brief—subject, location, action, lighting, lens, and mood.
Animate a first frame and optionally guide the ending with a last frame for a more deliberate transition.
Use a reference clip to communicate motion, rhythm, and visual intent. MP4 and MOV are provider-ready.
A practical creative workflow
Start with the simplest input that expresses your idea. Wan 3.0 Video handles the technical translation while you retain control over format, timing, and sound.
Generate from text alone, upload first and last frames, or provide a reference video up to 15 seconds.
Write the action first, then add camera movement, lighting, pace, and atmosphere for more intentional results.
Select aspect ratio, audio, duration, and batch count. Text and image generations support 5–30 seconds.
For consistent results, keep one clear subject and one primary action per shot. Put the most important movement near the beginning of your prompt.