CivArchive

    For on-site generation, I recommend only using models designed for WAN2.1 or earlier. It does not appear that Civitai can get 2.2 models to work. You can use the Universal model with both high and low noise steps, but if you want the best quality, I included both versions as well.

    This LoRA is trained on conversational language. You can describe your scene in plain English or Chinese, and it should adhere to your prompt well. Below are some inspirations for you if you need it.

    Positions:

    • On her back

    • Kneeling

    Perspectives:

    • POV

    Enhancement phrases:

    • She bounces her body up and down

    • She bounces her breasts up and down with her hands

    All showcase videos have the workflow metadata embedded. You can download the video and drag it into ComfyUI to replicate them with their prompts, seeds, and settings.

    Description

    Trained on a mix of new datasets and generated images from the previous generation, with some editing/modifying to prevent the model from training faces or other unwanted things. 5e-5 for 50 epochs, then increased the tensors in blocks by 20% with a python script to enhance movement (not a science, just tried a bunch of combinations of blocks and this worked out the best).