[LoRA] sh4g0d

模型介绍：
Hi!

"Presenting "sh4g0d". This LORA uses a different approach than the previous one but turned out really good (I think).

This LORA will bias your generations towards a cybernetic/horned style. Works very well for animals mixing (with some "advanced prompting"), works very good for simple portraiture prompts and even it works at 1 with some degree of variations, it shines at values 0.6-0.8.

Not recommended for objects or isolated gadgets but works fine with the right prompts. I made some crystal jar, glasses, beds, cars, etc. But the intention is more for characters.

Best usage is:
- Not using "sh4g0d" keyword at first (the LORA is very strong but without being overfitted) and you can decide if use sh4g0d (previously was another unreleased LORA that acted as "mechanizer", that's why it acts this way) or not to power your costume/char, etc.

IMPORTANT!

I wrote an article here and it's like my finetunning diary for the ones interested with releases upcoming, trainings done, exclusive patreon models previews, personal thoughts, etc. So if you want to see what I'm planning or what I'm into, feel free to take a look.

If you enjoy these LORA please check our exclusive content in Patreon where we train more styles like this one and some more complex, custom and experimental things:
https://www.patreon.com/user?u=90661954

Thanks a lot for your support and go crazy with this thing!

~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数：
Prompt: <lora:sh4g0d-000008:0.7> shaman, wearing mask made of wood, portrait

Negative prompt: easynegative

Steps: 14, CFG scale: 5, Sampler: DPM++ 2S a Karras, Seed: 1684609912, Size: 512x768, Model: revAnimated_v122, Model hash: 4199bcdd14, Hires steps: 15, Hires upscale: 2, Hires upscaler: R-ESRGAN 4x+, Denoising strength: 0.4, Clip skip: 2