Skip to main content

​
Supported inputs and limits

  • File types: MP4, MOV, or WebM.
  • Video length: 4 to 90 seconds.
  • File size: up to 500 MB.
  • Resolution: the shorter side must be at least 360 pixels; 720 pixels or more is recommended.
  • Target frames: 9:16, 16:9, 1:1, 4:3, 3:4, or 21:9.
AI Expand works best with camera footage and natural scenes. Generated areas can vary between results. Review important faces, objects, logos, and text before publishing.

​
How to use AI Expand

The AI Expand creation screen with Target frame, Model, and Resolution controls on the left and the upload area on the right. AI Expand controls and upload area before a source video is added.
  1. Open AI Expand. You can also go to Opus Labs, find AI Expand, and select Try it now.
  2. Drop a video into the upload area, or select browse to choose a file.
  3. Wait for the upload and video analysis to finish.
  4. Under Target frame, select the aspect ratio you want to create.
  5. Under Model, select Standard — Hailuo 3 or Premium — Seedance 2.5.
  6. Under Resolution, select an available output resolution.
  7. Review the credit cost shown on the action button.
  8. Select Expand to [target frame]. For example, select Expand to 9:16 for a vertical result.
  9. Wait for processing to finish, then preview the result.
  10. Download or share the result if it looks correct.
Animation of the AI Expand workflow: selecting a target frame, model, and resolution, uploading a clip, and submitting the job. Choose a target frame, model, and resolution before uploading and submitting.

​
Find a result later

Select Library in AI Expand to view submitted jobs and completed results.

​
Credits and availability

AI Expand uses credits. The cost depends on the selected model, resolution, source duration, and target frame. The exact estimate appears before you submit the job. Available models and resolutions can change while AI Expand is in beta.

​
Troubleshooting

​
The action button is disabled

Confirm that a video has finished uploading, then select a target frame, model, and resolution. The target frame must be different from the source video’s aspect ratio.

​
The upload is rejected

Confirm that the file is MP4, MOV, or WebM; 4 to 90 seconds long; no larger than 500 MB; and at least 360 pixels on the shorter side.

​
The generated area looks incorrect

Try a different model or use a source clip with a clear subject and less camera movement. Always review generated faces, hands, text, logos, and objects.

​
The job is taking longer than expected

You can leave the creation screen after submitting the job. Return to Library to check its status.

​
Frequently asked questions

​
Does AI Expand crop my source video?

No. AI Expand keeps the original frame and generates additional visual area to reach the selected aspect ratio.

​
Can I select the same aspect ratio as my source video?

No. Select a different target frame. AI Expand is designed to create a new aspect ratio.

​
Can I close the page while a job is processing?

Yes. After the job has been submitted, return to Library to check the status and open the result.