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:42 images, 15 steps, 10 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 0.3-0.8 (best 0.5)</li><li>Keys: wearing sexy maid dress</li></ul>This model is very easy to use, just use the keyword and if necessary give to it a weight. If you don't specify the color it will be black and white as the majority of the images of the training. but you can start your prompt with something like "a woman, (wearing sexy maid dress:1.2), red dress" to choice a particular color.You might also want to try different weights for the lora, i used 0.5 and it worked great. with higher weight it may be more accurate in the dress but you risk to get artifacts in your images or wierd faces.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: 2feb5e3e6c8114f3f0c4ddc0d57b13c80422c8790d25abdd76418923bab837d5