Based on the latest Wanxiang 2.1 720P model core, it directly produces high-quality 1080P videos in 6 seconds, allowing separate control of character actions and camera movements.

Using reverse prompt (more accurate character consistency and smoother actions), you can configure it according to the instructions below to output videos longer than 10 seconds.

Frame settings instructions: 123456 seconds

Note: 161 frames use processor region adjustment to set the original generation height and width limit to 512. If the original limit is set to 768, the maximum output is 73 frames, which equals a 6-second video. Beyond this, errors occur, but increasing the block swap value to 40 can resolve this. It is recommended to output a 5-second video, which equals 61 frames for the best results.

Final output dimensions: Regardless of the input image size, the final result will be centrally cropped to this size.


WanVideo BlockSwap: With the same number of STEP iterations (10), the larger the block swap, the higher the resolution that can be generated, but the time will be slower. Normally, a 6-second video with a 768 original image limit and a setting of 10 can run. If set to 1024 or even 1280, it is recommended to increase the block swap to 30 or 40 to avoid memory overflow. It is also possible to increase the duration at the same resolution but requires increasing the block swap size.

High-definition settings: Original image limit 768, output arbitrary, iteration 30, block swap 10, maximum video duration 6 seconds (73 frames), generation time (10 minutes)

High-quality settings: Original image limit 720, output arbitrary, iteration 20, block swap 10, maximum video duration 6 seconds (73 frames), generation time (8 minutes), can attempt 7 seconds (85 frames)

Medium-quality settings: Original image limit 512, output arbitrary, iteration 10, block swap 10, maximum video duration 6 seconds (73 frames), generation time (5 minutes), can output 10 seconds (121 frames), maximum 161 frames, 13.5 seconds (tested)

Low-quality settings: Original image limit 384, output arbitrary, iteration 10, block swap 10, maximum video duration 6 seconds (73 frames), generation time (3 minutes), can attempt longer durations