If you do like this work, consider buying me a coffee :)
Use v6 of this model for free on Stable Horde
Use this model for free on Shakker.ai
The long awaited followup to ICBINP, this model is merged to try and get the most realistic images possible. It's the ICBINP you all know and love, but now with all of the benefits of SDXL (recommended 832x1216)
Update: v6
More mixing to try and get something even better than v2, and I think I've found a much nicer balance
The sample images are made in Forge with the DPM++ 2M Karras, 35 steps, 832x1216, cfg 3 - 4, clip skip 2, (some hi-res fix, but the best realism is without it IMO - use the Ultimate SD Upscaler instead) and no additional tooling so should be reasonably easy to replicate.
If you want to push this to make anime, change the sampler to DPM++ SDE 3M Exponential and crank the CFG up to 6-7.
I'm sure that some more experimentation and better workflows you will be able to achieve amazing results!
Description
Custom block weighted merge of Jib Mix, Juggernaut, SDXL base, and the add detail xl lora
FAQ
Comments (10)
What are the models used for the merge?
Jib Mix, Juggernaut XL, SDXL base (all in version info)
@residentchiefnz Thanks, I wasn't sure whether it's the full list.
Think we can use kohya dreambooth training on this as base? to use with our faces\models?
I cant see why not
what are your hires fix settings if I may ask? what sort of step count and denoising strength are you using with R-ESRGAN 4x+, and have you manually altered the tiling settings at all? how far are you upscaling? 2x?
Upscale was 10-15 steps at 0.3-0.4 scale by 1.5
@residentchiefnz tyvm, interestingly that confirms what I already normally use, I do almost the exact same usually, though I use more steps when i probably don't need to
@shapeshifter83 No worries :)
I have done several tests and comparisons with other models on Fooocus and at the moment this is the one that seems to be the most photorealistic and also at the same time very creative, congratulations, I can't wait to see updates