One model for text, first and last frames, reference images, and source video.
A 5-second generation is 150 credits at 480P, 250 at 720P, or 450 at 1080P. The composer shows the exact cost for your selected inputs.
Open the generator when you are ready to upload a reference or write a prompt.
Open generatorThese are existing vid2vid previews used to illustrate source and output workflows. They are not Wan 3.0 benchmark outputs.
Start with a clip when the timing and movement should remain recognizable while the visual direction changes.
Transform the city walk into a cinematic blue-hour scene while preserving the subject, walk cycle, framing, and camera path.
Use an image or reference set to explore a clean product treatment, controlled lighting, and measured camera motion.
Create a premium technology commercial with clean reflections, precise edge lighting, stable product detail, and restrained motion.
Choose the input that matches your starting material. The unified vid2vid composer exposes the supported controls without inventing separate model variants.
Describe a new scene when you do not have an image or clip to start from.
Provide a first frame, or both first and last frames, to guide the shot transition.
Use up to 10 reference images and 5 reference videos within the composer limits.
Generate 2-30 seconds at 480P, 720P, or 1080P. With video input, input plus output duration must stay within 30 seconds.
The model-specific evidence is the connected KIE contract for wan/3-0-video. Shared vid2vid previews illustrate workflow shape only and are not Wan 3.0 output samples.
Model id: wan/3-0-video.
Inputs exposed on vid2vid: text, first frame, first and last frames, reference images, and reference video.
Output: 2-30 seconds at 480P, 720P, or 1080P; video input plus output must be no more than 30 seconds.
Wan 3.0 is useful when one workflow needs more than a single input type.
Turn a written shot idea into a short scene for concepting, previsualization, or social content.
Combine frames and references when the product, subject, or visual identity must remain guided.
Start with a video and describe a new look while respecting the documented combined-duration limit.
The generator adapts to the media you attach and keeps the final request in the Wan 3.0 input format.
Start with a prompt, add a first or last frame, or attach a source clip. Reference controls are available when the selected input supports them.
Use up to 10 images and 5 videos. Reference videos can be 1-15 seconds each and 15 seconds total.
Choose 2-30 seconds and 480P, 720P, or 1080P. Review the credit estimate before submitting the task.
Supported inputs, output limits, references, and pricing.
Start from a prompt, frames, references, or a source clip. The generator displays the applicable controls and price before creation.
