<h2 id="konosuba-show-style">Konosuba - Show Style 📺</h2>This LoRA on my screen shots of my favorite anime show [<a target="_blank" rel="ugc" href="https://anilist.co/anime/21202/KONOSUBA-Gods-blessing-on-this-wonderful-world/">KONOSUBA -God's blessing on this wonderful world!</a>], i mainly targeting the expressions and scenery by training this so its something between a style and a concept i think.If you didn't watch the show already consider giving it a try it's super funny 💚<hr /><h3 id="usage-tips-(il)-:">Usage Tips (IL) :</h3><ul><li>I personally use it at a weight of 1.1 so i recommend that</li><li>the LoRA needs to have its trigger word "<code>nv-st-kono, anime coloring,</code>" in the prompt<ul><li>in case you wanna mix it with other named artstyles effectively then remove "<code>anime coloring,</code>"</li></ul></li><li>so far only tested on <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184?modelVersionId=1761560">WAI-NSFW-illustrious-SDXL</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184?modelVersionId=1761560">V16</a> (used in the showcase pics)</li><li>preforms well well on male characters</li></ul><hr /><h3 id="training-notes-(il)-:">Training Notes (IL) :</h3>used a manually curated dataset around 200 pics with the following traits:<ul><li>90 pics main characters / 120 pics non-main characters</li><li>30% / 70% male to female ratio </li><li>10% non human pics data backgrounds, animals</li><li>most data cropped to focus on the main subject</li></ul><hr /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6b9fae0b-64d2-497a-a017-246be41d2526/original=true/6b9fae0b-64d2-497a-a017-246be41d2526.jpeg" />

Files with SHA256: d7f6d732513021be814c3ccefd1a47b823d625895593be96675e62321028b5f5