<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/31f745fb-aa84-4ccf-e986-af380b6bbc00/width=525/31f745fb-aa84-4ccf-e986-af380b6bbc00" /><h3>What's this LoRA</h3>This is a LoRA trained on a particular style. This LoRAs has a nice 2.5d syle with soft colors and borders.The images in the examples are created using only this LoRA and different models (mainly my customs: <a target="_blank" rel="ugc" href="https://civarchive.com/models/24149">Mistoon_Anime</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/26332">Mistoon_Ruby</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/28322/mistoonemerald">Mistoon_Emerald</a>)You don't need to specify a word to trigger this LoRA.The LoRA works well on NSFW (check last examples).The images are created using different weights (you can check on the image generation panel)<h3>Different Flavors</h3>My LoRAs are usually trained on AI generated images in two different steps. The first training uses images generated using a model (I choose it based on what I'm going to build) with a size of 512x768, and this is usually 10 images with different content but a similar style. The second training is based on pictures generated using the LoRA from the first step. This is done at a higher resolution (768x1152) and the output is usually upscaled using img2img.The LoRA is available in two different flavors:<ul><li>RAW: This is based on the first batch of images, it has usually a more creative output, but usually needs hiresfix or img2img to really shine</li><li>Standard: This is based on the secon batch of images, it has usually a less creative output, but way more stable.</li></ul>Here's a quick comparison (not cherrypicked):RAW:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/00ddcc9d-3835-4a3c-ca8e-0edc354b0100/width=525/00ddcc9d-3835-4a3c-ca8e-0edc354b0100" />Standard:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/24d7201c-0eb6-4527-5ea9-72a96d08a700/width=525/24d7201c-0eb6-4527-5ea9-72a96d08a700" /><h3>Guides</h3>Learn how to create pictures like mine with my step-by-step tutorials:<ul><li><a target="_blank" rel="ugc" href="https://medium.com/p/bd7dbcd5ce4b">Beginner's Guide</a></li><li><a target="_blank" rel="ugc" href="https://medium.com/p/35eacb3dc5f4">Prompting</a></li><li><a target="_blank" rel="ugc" href="https://medium.com/@inzaniak/stable-diffusion-ultimate-guide-pt-3-high-resolution-a4f5d7b60f38">My Workflow</a></li><li><a target="_blank" rel="ugc" href="https://medium.com/p/772ea69472c9">Inpainting</a></li></ul>If you want to create images with a high LoRA weight, I highly suggest you to check the tutorial.<h3>Support</h3>If you want to support my work, you can check out my music here:<a target="_blank" rel="ugc" href="https://inzaniak.bandcamp.com/">https://inzaniak.bandcamp.com/</a>You can also check my AI Art accounts, where you'll find daily pictures made with my models (and previews of new ones coming):<ul><li><a target="_blank" rel="ugc" href="https://www.instagram.com/inzaniak_aiart/">Instagram</a></li><li><a target="_blank" rel="ugc" href="https://www.deviantart.com/inzaniak">DeviantArt</a></li></ul>

Files with SHA256: ba168c49cf57c9ce3783f5c8d2882a3b46f67464d0a47486bbbe514d8a78d5fb