| Check | Standard test | HD test |
|---|---|---|
| Final resolution | 480 × 854 pixels | 720 × 1280 pixels |
| Final video duration | 8.000 seconds | 8.000 seconds |
| Frame rate | 30 fps | 30 fps |
| Audio selection | BGM off: 0 audio tracks | BGM on: 1 AAC track |
| BGM verification | Silent file | Encoded audio packets matched the fixed soundtrack |
| Provider processing time | 610 seconds (10 min 10 sec) | 854 seconds (14 min 14 sec) |
What should I compare in the two public videos?
Look at the main dancer’s face, clothes, hands and turns. The original shows the party and camera movement; the supplied AI example shows a cowboy as the foreground subject. Background people and small movements may differ. Your photo should guide the appearance and clothing in your own result.
Was the supplied cowboy example generated by Dime Baby?
The cowboy clip was supplied with the project. Its original generation model has not been independently verified. New Dime Baby jobs use Alibaba Wan 3.0 through Kie.ai. The 2 measured preview outputs in the table are separate acceptance samples, not the supplied showcase clip.
What did the 2 generation checks prove?
Both downloaded MP4s were 8.000 seconds and decoded successfully. The Standard test had no audio track; the HD test had the exact fixed BGM packets. These checks confirm the two tested files’ duration and audio handling. Two samples do not establish a success rate, likeness guarantee or typical processing time.
How do I hear the example soundtrack?
Use the AI example’s video controls to enable sound. Homepage autoplay starts muted. The supplied original clip has no audio track. For your own generation, leave Include original BGM on to add the example soundtrack to the final MP4.
More questions about this topic
Are the test results a guarantee for every uploaded photo?+
No. They describe 2 files checked on 21 October 2026. Faces, hands, clothing and motion may contain errors even when a file passes the duration and audio checks.
Why is the original longer than the AI example?+
The original is preserved in full at 9.17 seconds. The AI example and fixed motion reference use only their first 8 seconds to match the requested output length.