<a target="_blank" rel="ugc" href="https://beacons.ai/diffusiondudes"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c5d38298-efb3-4fc9-8e8d-cc6f2dec38d8/width=525/c5d38298-efb3-4fc9-8e8d-cc6f2dec38d8.jpeg" /></a><a target="_blank" rel="ugc" href="https://beacons.ai/diffusiondudes">Visit my Patreon/Gumroad/Ko-fi for exclusive access to all 80+ of my LoRAs! ▼▼▼▼▼</a>Members can get full access to my entire archive plus my Discord server for my steamiest content! 💪-From the front AND the back! 🍆 🍑 (though the back works way better)I have been sitting on these for a minute and frankly almost forgot I made them until a bounty was posted. An ode to the guys I grew up around who refused to keep their pants up. I divided this into two separate LoRAs, front and back. The front definitely isn't as easy to reproduce as the back. But the back is the best part in my opinion :) This is a LoRA trained on the base 1.5 model, at 768px. Please share your results and feedback for any future training improvements.Recommendations (Front):<ul><li><code>Trigger token: sagfrnt, sagging pants</code></li><li>Works best on photorealistic models, but might work well on others (I haven't explored)</li><li>Clip skip: 2 works better for some reason sometimes</li><li>Strength: 0.8 - 1.2 (I recommend using it at 1 and increasing the trigger token weight)</li><li>Works best with clothing descriptors (sagging pants, sagging shorts, boxers, etc.)</li></ul>Recommendations (Back):<ul><li><code>Trigger token: sagback, sagging pants</code></li><li>Strength: 0.8 - 1.2 (I recommend using it at 0.9 and increasing the trigger token weight)</li><li>Works best with clothing descriptors (sagging pants, sagging shorts, boxers, etc.)</li></ul>I am not responsible for your use of this resource, which is not for commercial use and is for research purposes only.

Files with SHA256: c871115c2a0ea937d292e0b461b14a92591c3876baed2779e15be55ff698c20a