Hollow knight [Wan 2.2 T2V-A14B]

https://civitai.com/models/1836374/hollow-knight-wan-22

Trigger Word: hollow_knight
Model: Wan 2.2 t2v A14B
All examples are generated with 1.0 strength LOW + HIGH LoRA
For inference used tazs-example-workflows

Workflow Details

I used both the default workflow and the lightning-4-step workflow. (Thanks a lot to tazmannner379 for sharing worflows.) Somewhere I read that in some cases, lightning can make the result better than a "simple" wan, but I didn't expect them to be SO MUCH different. Not only does inference happen MUCH faster with the use of lightning-4-step, but the generation itself has become more saturated and brighter, and a more noticeable cartoon style has appeared, just like from the game. So I recommend using it.

Perhaps this is due to the fact that the training dataset itself is very dark. Therefore, the lightning LoRA infers more extra information than a "simple" wan. A side effect is that it is difficult to generate a dark video using lighting. No wonder this lora is called Lighting hehe.

Gallery

I didn't really want to upload the results to the gallery with only the "high" or only the "low" LoRA. So all you see in the gallery is LOW + HIGH LoRA. The only difference is that a "simple" wan is used for the HIGH section. And for the LOW section lightning-4-step. Maybe it's a little confusing, but I wanted to show both launch options for this LoRA.

Dataset Details

Information about the dataset is described in the previous article.