<a target="_blank" rel="ugc" href="https://my.bio/diffusiondudes"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2b56a416-6ad5-4f3f-8334-08e0a9d735e3/width=525/2b56a416-6ad5-4f3f-8334-08e0a9d735e3.jpeg" /></a><a target="_blank" rel="ugc" href="https://my.bio/diffusiondudes">Visit my Gumroad/Ko-fi for exclusive access to all 80+ of my LoRAs! ▼▼▼▼▼</a>I'll be releasing 1-2 LoRAs on Civit every month, with early access on my stores, and over 80 exclusive models 💪 Members can get full access to my entire archive plus my private TG for my steamiest content!-The Prince of Dumbass Kingdom! My first go at a Sanji LoRA, this works okay but really wants to do low angle fisheye closeup shots because that's what the majority of my data was. Hopefully this will be updated with a second season. I also included some data of Taz Skylar outside of the character to help alleviate the issue, so you might need to define his hair. This is a LoRA trained on the base 1.5 model, at 768px. Please share your results and feedback for any future training improvements.Recommendations:<ul><li><code>Trigger token: sanjiop</code></li><li>Getting his hair to cover one eye is super dependent on pose and model, add 'hair covering one eye' might help</li><li>He'll come out with more facial hair than he actually has, use 'facial hair' in negative prompt</li><li>Strength: 0.7-1, I usually start at 0.8</li><li>I recommend using After Detailer + Hires Fix, or IMG2IMG double upscaling using MultiDiffusion and ControlNet Tile for highly accurate faces</li></ul>I am not responsible for your use of this resource, which is not for commercial use and is for research purposes only.

Files with SHA256: 59928c01adbdb9c805055dacdd63549003dfe23430ad6bd426731230b24d812b