(see license restrictions below)
What is this thing?
MoP Pro is something I've been working on for a while. The base for this model is the latest version of my MoP (Mixture of Perverts) model. I've done a significant amount of training with various datasets I've compiled. These datasets contain 500 images and have been generated with my GonzaLomo ZPop Insta and MoP Mix Big ASP 2.5 models. I really enjoy the aesthetic qualities of those models and wanted to try to replicate them in a fast. normal SDXL model.
MoP Pro is not going to provide the highest detail or the crispest image, but that is by design. The focus on the training is in overall image composition, lighting, facial diversity/expressiveness and a certain everyday photographic quality. It's one of those things that's not easy to explain but I hope the example images I've provided do the explaining for me.
The recommended settings are:
Sampler: LCM
Scheduler: karras, beta, beta57, bong_tangent, ddim_uniform and exponential
Steps: 6-24
CFG: 1.0-1.7
It's very flexible with steps. In general, the lower the steps the higher the creativity but the lower the pixel quality. The reverse is true when you raise the steps. I've gotten good results with anywhere from 6 to 24 steps. bong_tangent really needs at least 16 steps and a cfg of 1.4 at least though.
It also isn't restricted to 1024x1024 sized latents. I reguarly use 1152x1536 for 4:3 ratio and 960x1728 for 9:16.
MoP Pro License
1. You may use MoP Pro non-commercially without crediting the creator.
2. You may not sell the images that it generates without obtaining a separate license or permission from the creator.
3. You may only run the official MoP Pro on Civitai. Running it on other public or shared servers commercially requires a separate license or permission from the creator.
4. You may not sell this model or merges of this model.
5. You may not merge, share merges of, or share LoRA extractions incorporating MoP Pro.
6. Redistribution of MoP Pro is prohibited. It may only be obtained from official sources designated by the creator.
7. Any restrictions in these additional terms may be waived only with explicit written permission from the creator.
Description
Some additional training with private datasets. Hopefully a bit better definition with eyes and teeth while still retaining some of the improved composition and color.
The recommended workflow is https://civitai.red/models/1513601/gonzalomo-sdxl-workflows?modelVersionId=2968737.
FAQ
Comments (12)
MoP Pro 2.1 great stuff ! I love it ....
@GBRX >>You may only run the official MoP Pro on Civitai.
That includes offline/local use if downloaded from CivitAI, right? Otherwise, I'm not sure why the download option is enabled. Anyway, just posted a locally-generated sample result purely for educational/testing purposes, so please let me know if you want it deleted or anything (I don't always read or understand licensing these days, but that jumped out at me).
No, that just means that CivitAI is the only online generation service it can be run on. You can certainly run it on your own personal machine. Maybe that could be worded better.
best model I've ever tested. i am new, i have less than 2 months, and honestly, I don't know whether to consider it, a good or a bad thing that I started with this one. Highly recommended
Thanks
really appriciate your great work. been using all versions of your models.
could we this this version in Non DMD for training your own character loras? OR have anyone gotten good results training character loras on DMD baked in models?
Hey, thanks! I don't recommend ever training a lora on a DMD model though - it turns out pretty bad in my experience. You could try to remove the DMD-ness by adding the DMD lora with a -1.0 strength and see what happens. However, I've trained this model with a lot of my own images and I'm not sure how well it would work for lora training.
@GBRX The goat! not only the best models, Also fast responses.
Yeah i figured! Ive trained on almost all mainstream non DMD checkpoint merges and yours (MOP v.5.1 NON-DMD) is by far both best checkpoint but also takes identity extremely well compared some others!
if you ever feel to bless the community further, would greatly appreciate non dmd baked versions of This or any of you models TBH
Another great model, thanks. But with this one how to avoid skin tanlines ?
That can be tricky. Sometimes 'pale skin' or 'dark skin' can help
is DMD2 baked into this model? I don't see that mentioned anywhere but it seems to be implied by some of the comments
Yes, it is a DMD model.