How do I create a video and find its result?

Use the Video workbench for video generation. Available text, frame and reference-material options depend on the selected model.

1. Open Video and choose an available model and input option. Enter your description and add the material required by that option.
2. Wait for material uploads to finish. Check the available aspect ratio, resolution, duration and audio settings.
3. Review the displayed cost and generate. Check the generation status in the project results.
4. When complete, use the result’s available playback and download actions. Reopen the project from Video’s project records when needed.

For a problem, tell Support the model, input option, project, time, visible state and failing action. A processing item is different from a completed video. Do not submit again merely to check an existing generation. Supported inputs and output settings follow the current model options.

Video generation requires a paid plan; having characters left is not enough on its own. In Video, you can select a finished Image result from the same account without downloading and uploading it again. Start/end frames and reference images are separate input modes and cannot be combined; add a start frame before an end frame. Whether you can change the duration with reference images depends on the model. It is not always eight seconds. After submitting, use the history link in Video to find the original project.

The current Video model options include Google’s Veo 3.1 Fast and Veo 3.1, and ByteDance’s Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini and Seedance 2.5. You can view and switch models in the Model dropdown in Video. Input modes, resolutions, durations and native audio options can differ by model. Use the options shown after selecting a model.