<h2>Ahri (League of Legends) LoRA</h2>Making models can be expensive. Do you like what I do? Consider supporting me on <a target="_blank" rel="ugc" href="https://www.patreon.com/Lykon275">Patreon</a> 🅿️ or feel free to <a target="_blank" rel="ugc" href="https://snipfeed.co/lykon">buy me a coffee</a> ☕ Requested by plsdontultme. As always, this is a LoRA, so you need to use it as additional network on top of your main model. For the example I set the weight to 0.6. Activation words are <code>ahri</code>, <code>ahri \(league of legends\)</code>, <code>k/da \(league of legends\)</code>, <code>spirit blossom \(league of legends\)</code>. Yes, it works on alternate costumes too. If Ahri doesn't show up, try to add <code>animal ears, facial mark, fox ears, fox tail, orange eyes, multiple tails, tail</code> to the prompt. As for the model I suggest AbyssOrangeMix2 or AnyLoRA.How to use LoRA's in auto1111:<ul><li>Copy the file to <code>stable-diffusion-webui/models/lora</code></li><li>Click on the "show extra networks" button under "Generate" (purple icon)</li><li>Go to the <code>Lora</code> tab</li><li>Click on the net you want to apply, it will be added in the prompt</li><li>Make sure to change the weight (by default it's <code>:1</code> which is usually too high)</li></ul> 45 epochs seem the sweet spot for this dataset. If the 45 version is too strong, lower the weight or grab the 30 epoch version. If you find it to weak even at 1.0, try the 50 epoch version. The 30 epoch one is also a bit more sfw.

Files with SHA256: d08eb8da757352418dd47443085f7f41d3f8d513237b07072f2511e0e5145121