Use this LoRa to open up a full new dimension to your images. Increase slowly the strength and see the magic happen.
Prompts should have the word 'art,' as first one. The dataset was captioned like '..captivating watercolor drawing of a womans face with exiciting make up. The focus is on her striking..' but bland prompts like 'art, female face' do work as well.
On the base SDXL model the strength can be from 0-1.5 for a lot of themes, i use it with success on a custom model mix from -1 to +3 with interesting results.
Have fun and enjoy, this is already v2.
It would be nice if you share your creations here, i am curious how people are going to use it.
Thank you <3
Description
The fourth iteration of the Nicify LoRa lets you explore another dimension in your faboulous creations. Wether you're accustomed to employing AI for elaborate prompts or writing them on your own, this LoRa is tailor-made for you.
In this latest version, while you can still utilize the familiar trigger word 'art', it's no longer a necessity. With a lot more images integrated, the lengthier your prompt, the more pronounced the enhancement. Again, everything is up to you.
Fine-tune the strength parameter anywhere between 0 and 1.2, occasionally pushing boundaries up to 2 for exceptional results.
The LoRa is trained on the base model, but i highly recommend something more finetuned and let the LoRa boost the inherent variety.
FAQ
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.