Request a Wan2.1 LoRA on our Discord and we will train and open-source it for free.

Join our Discord to generate videos of the Decay Effect LoRA for free.

Wan2.1 14B I2V 480p v1.0:

Trained on 30 seconds of video comprised of 6 short clips of things timelapse decaying. This was trained on the Wan.21 14B I2V 480p model.

The trigger word is: 'd3c4y decay time-lapse'

See below for the prompt structure that's worked best for me.

Recommended Settings:

LoRA strength = 1.0

Embedded guidance scale = 6.0

Flow shift = 5.0

Here's a link to the Wan2.1 I2V LoRA inference workflow I used to generate these videos: https://huggingface.co/Remade/Squish/blob/main/workflow/wan_img2video_lora_workflow.json

This is a slight modification to Kijai's version, with the main difference being the addition of the WanVideo Lora Select node, connected to the 'lora' field of the WanVideo Lora Select node. Find Kijai's original workflow here:
https://github.com/kijai/ComfyUI-WanVideoWrapper/blob/main/example_workflows/wanvideo_480p_I2V_example_02.json

Prompt Structure:

The video shows a [object]. The d3c4y decay time-lapse begins, causing the [object] to change. The [object] is initially whole, but soon it appears to be rotting. The [object] slowly becomes increasingly shriveled and discolored, and eventually, the [object] decomposes and falls apart. The [object] is rotting in the center and appears to be covered in mold, completing the d3c4y decay time-lapse.

Let me know if there are any questions, I'll be happy to help!