Get early access to build and test build, be able to try all epochs and test them by yourself on <a target="_blank" rel="ugc" href="https://www.patreon.com/betterCuteAsians">Patreon </a>or contact me for support on DiscordTo support me you can buy me a coffe at <a target="_blank" rel="ugc" href="https://ko-fi.com/antonioriolo">https://ko-fi.com/antonioriolo</a>Training settings:40 images, 15 steps, 12 epochs, you find many epochs (with test images using plot xy script) on my <a target="_blank" rel="ugc" href="https://www.patreon.com/betterCuteAsians">Patreon</a> while supporting me. Recommended settings:<ul><li>Make sure you are aware on the usage instructions of LORA</li><li>Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)</li><li>CFG Scale : 5 to 9</li><li>LORA weight for txt2img: reccomended 0.4-0.7 (0.4 best results in private environment, 0.5-0.6 in public environment)</li><li>Keys: wearing see trough dress, elegtant dress, lingerie, back view</li></ul>this model isn't hard to use, just use the keywords you need.An advice i want to give is that if it's too much dressed and see trough isnt working you might want to place the keyword "see trough dress" at the beginning of your prompt and give to it an higher weight. Instead if it's too much naked you should do the opposite, placing it at the end or in the middle and/or lower his weight. As much as i tryied, see trough work well in private scenes, but when in public you really need an higher weight to the keyword or she will be dressed without see trough. Enjoy! If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: <a target="_blank" rel="ugc" href="https://www.patreon.com/betterCuteAsians">https://www.patreon.com/betterCuteAsians</a>

Files with SHA256: ad2f97d3566f104f596046c5d02c7ba35b39cb39d6a372db47fa3d79f90d625b