🎞 LoRA is trained purely on anime screencaps
🗝 Activation tags are in the right panel
📸 Always use ADetailer to get detailed eyes
🛠 Technical notes:
LoRA shoud be compatible with Illu models & NoobAI. (Not tested.)
LoRA is compatible with various styles - Lower the weight & remove "anime screencap" from the prompt, if you are getting too strong anime style
Hi. res fix used on all sample images:
Upscaler: R-ESRGAN 4x+ Anime6B
Steps: 10-15
Denoising: 0.3
📬Leave a like, comment, review but first of all, ENJOY ;)
Description
FAQ
Comments (10)
hi, could you open the permission to download? thanks very much!
fixed.
@LittleJelly thanks!
Question, why any generation is so yellow? like its sunset, mostly yellow tone colors any prompt to fix colors so they look like the ref images?
Huh? Might be the checkpoint issue? I'd need to see your prompt & the resulting image with all loras used.
Also you might be missing some basic illu negatives?
Although the staff in your images is clearly Frieren, in reality it turns out to be a Fern staff, although I needed to further strengthen it.
I have no idea about what you just said. No heads or tails. Elaborate more.
Frieren what? Fern what?
@LittleJelly Staff https://civitai.com/images/54157053 Frieren not Fern
but I can get the Fern staff https://civitai.com/images/58197874 although I had to reinforce his prompts wooden staff,
@fansay oh that... It's not about fern or frieren that's just "holding staff" tag from every checkpoint. There's no frieren staff inside the dataset. I didn't train it specifically for fern staff either so you can try putting "gem" inside the negatives.
10/10 🔥
Details
Files
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.