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 Fubuki Wan2.1 LoRA for free.
Hey everyone, really excited to be sharing this, my first Wan2.1 14B txt2vid LoRA! Very recently finished training it so haven't had too much time for testing, and I will be waiting for the paper to come out so we have a better idea of what the best config parameters are. In the meantime, you can join our Discord to generate videos using Wan2.1 for free, and also request LoRAs for it! So far I've managed to train Hunyuan Video LoRAs for almost all requests, and I'll do the same for Wan2.1!
Wan2.1 v1.0:
Trained on Wan2.1 14B txt2vid model with 13 images of Fubuki in various episodes of the One Punch Man anime.
The trigger words are: F08uk1_0P8, Fubuki, green hair
Recommended Settings:
LoRA strength = 1.0
Embedded guidance scale = 5.0
Flow shift = 6.0
Prompt Recommendations:
Best to keep the prompt relatively simple, here are some examples that have worked well for me:
Example 1: 'F08uk1_0P8, Fubuki, green hair leans against a dimly lit streetlamp in a black-and-white city, exhaling as glowing green smoke swirls from an old-school cigarette.'
Example 2: 'F08uk1_0P8, Fubuki, green hair floats weightlessly in an underwater abyss, bioluminescent creatures swimming around her as her psychic energy ripples through the water.'
Example 3: 'F08uk1_0P8, Fubuki, green hair walks forward in a snowy blizzard, her coat billowing dramatically as she extends a hand, psychic energy crackling around her.'
Let me know if there are any questions, I'll be happy to help! I'll start training more Wan2.1 LoRAs right away.