*New Version v1.1 has been trained on a larger dataset and produced some really good images. The lora file size is much larger due to the new training method that was used and the larger dataset. Check it out and let me know what you think! Do you like the original version better, let me know.WOW, BUBBLES!! This lora creates some amazing bubbles! I would suggest trying 3 different way to prompt it. Trigger word is the same as the file name so it's easy to remember, wow_bbls.7 on the LoRA strength seems to be about the best..<ul><li>a (subject) made of bubbles, wow_bbls</li><li>a (subject made of transparent bubbles, wow_bbls</li><li>a (subject made of iridescent bubbles, wow_bbls</li><li>a (subject made of colorful bubbles, wow_bbls</li></ul>Using "transparent bubbles" or "iridescent bubbles" or "colorful bubbles" will give you some great images. I was amazed I was able to get all those different variations for this lora! Let me know what you think of it in the comments below and please post your images you create with it, I want to see them!Special thanks to @toasters for creating the text image! Be sure to check out his YouTube channel for some great tips and trick using Fooocus! You'll be blown away on what it can do!Great text with Stable Diffusion -by <a target="_blank" rel="ugc" href="https://www.youtube.com/@KLEEBZTECH/">Kleebz Tech AI</a><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/CYP6SpllsK8?si=h3_c5Y8YwjE3UhZk" start="0"></iframe></div>

Wow Bubbles - SDXL Style LoRA

v1.1 has been trained on a larger dataset and produced some really good images. The lora file size is much larger due to the new training method that was used and the larger dataset. Check it out and let me know what you think! Do you like the original version better, let me know.

wow_bbls_V1.1.safetensors

v1.1

v1.0

&lt;p&gt;v1.1 has been trained on a larger dataset and produced some really good images. The lora file size is much larger due to the new training method that was used and the larger dataset. Check it out and let me know what you think! Do you like the original version better, let me know.&lt;/p&gt;

Files with SHA256: f0105ac38703eda461860d892c716630d32855d1d6b77b574716a2fa8ec1cd71