Zlixon style AnimaB1

The style of Zlixon.

The trigger word is @zlix0n. This is a Unet-only DoRA for Anima Base 1.0. It's trained on the same dataset as my Illustrious model of this artist and with the same tagging (except for the changed trigger word), so same remarks may apply.

This is my first time training for Anima -- I'd been delaying any training due to other preoccupation, but now that Anima is officially out of preview, I wanted to try to do something with it. I didn't have time for dataset prep, so I took one of my existing ones. I might redo this in the future if I find Anima requires different tagging, but so far the results seem satisfactory. I'm also still figuring out the optimal parameters and experimenting with new configurations. I decided to publish this and future different-model ports as separate models, as opposed to versions of previous ones, to avoid confusion. I also decided to go with opaque version names -- I used to use base model + epoch number, but I very rarely ended up uploading more than one epoch from the same training run, and more often uploaded versions from different runs. Once I have enough free time, I'll remake my previous LoRAs for Anima and add some new ones to the mix.

I'll also try not marking this as "intended to produce mature themes". I'm not entirely sure how CivitAI means this to be applied to styles. The dataset includes both SFW and NSFW artworks, and the model would've been capable of both anyway, even if didn't.

If you like the style, please support the original author.