Feel free to post, rate, follow, emote, or drop feedback! 🍻🙇♀️
Bootybear Style v3
I used WAI-illustrious-SDXL v15 for these images, feel free to try other checkpoints! Though certain checkpoint or LoRA(s) may heavily influence the style.
Generation details, prompts, and model info are listed with most images. CFG Scale 5 (5-7), 20-30 steps, and Euler a/DPM 2M as the sampler, Hires Remacri. Feel free to change.
Use ADetailer to help with eyes/eye lashes/face or try Hires fix. Due to style the model does have FREQUENT issues with these, this helps. Eventually, I'll try to see if switching to other training models helps, involving illustrious/noob.
Use Trigger Word: bxxtybear
...if you don’t, it won’t work well
Description
self-trained instead of on-site
removed older and "out-of-style" images
moved away from Prodigy to AdamW8bit
images are prompted way better, more improvements towards anatomy
for LoRA trainers, BIRME sometimes causes artifacts and noise issues when cropping large images, I swapped to photoshop and is much better at cropping and keeping quality
FAQ
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.