euler a with normal, or er sde with sgm uniform, 30-50 step, if using turbo 15 step fine most of the time,pls support if you like model <a target="_blank" rel="ugc" href="https://ko-fi.com/rastaai">https://ko-fi.com/rastaai</a>. i love making those lora but since i makin it on cloud gpu it'll be really useful for training, thx.

Sora no Iro, Mizu no Iro Hanime Style | rastandaa

soramizu_style_lora_v1-000010_clean.safetensors

v1.0

use it with sampler (euler ancestral and scheduler normal) or (er sde and scheduler sgm uniform) at weight lora around 0.7 to 1, with between around 40-70 step, if you're using turbo just 15 step it's still pretty good there is no trigger word for this model #################################### I accept LoRA training commissions for both Anima and Illustrious!If you're interested, want to make a request, or simply support me, please check my Linktree to access my Ko-Fi:🔗 <a target="_blank" rel="ugc" href="https://linktr.ee/rastaAI2">https://linktr.ee/rastaAI2</a>Even the smallest contribution means a lot and truly motivates me to keep creating models and tools for the community — especially since I'm currently training everything on cloud GPUs. 🤍Also, feel free to explore my model library — I love making LoRAs, though my page is still small at the moment. With your support, I'll keep making more in the future!

Files with SHA256: c1e1b77a2845c622a0c24ef730572e1028a98882ae249e0651dcd9300f775c18