Tetsuya Nomura style LoRA
Making models can be expensive. Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee ☕
I initially made this one only to further improve my Aqua (KH) pics, it turned out great. It can mimic Nomura's style incredibly well and has even picked up some characters (I managed to do Cloud and Tifa easily, without TI or other description besides 1boy / 1girl and their names).
Triggers with nomura tetsuya but it will pick up the style regardless of the trigger word. Example pics have been made on AnythingElse and AnyLoRA. I suggest weight 1 for the offset version (0.6 for the old one). Last 2 pics are also using my Aqua LoRA.
How to use LoRA's in auto1111:
Update webui (use
git pulllike here or redownload it)Copy the file to
stable-diffusion-webui/models/loraSelect your LoRA like in this video
Make sure to change the weight (by default it's
:1which is usually too high)
Description
FAQ
Comments (6)
nice work as always. i see you updating models with offset versions. i want to know ehat this offset version does and how to train an offset version. any info or tutorial you can give?
it's the same exact model as before. I just edit it so that you can keep it at weight 1. You can see from the comparison that it's exactly the same result, so it's not a retrain.
@Lykon how do you edit the model for weight 1? whats the process?
@Goofy_Ai kohya merger with square root of the sweet spot as ratio
any thoughts of making a flux lora for this?, i would make one myself if i weren't out of buzz lol
SD1.5 is outdated. Can it be refurbished and retrained to Illustrious or other model types?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.