Brought to you by @FallenIncursio GPUs That I made spin for hours because I was not liking the bakes it was doing.
This is my Luminaverse for pony XL. Trained on a handful of personal favorite images generated using MidJourney/NijiJourney.
The model is now available for Civitai Generator.
How to Proompt
Hold up!! Sadly I haven't yet finished writing a proompting guide which I started writing like 2 months ago. So here is the short of how to actually Proompt. Since Its Civitai User base we're talking.
This is Pony XL Model. (Important)
XL models don't require any "Omega LUL Schizo 100KM long random string that even I won't use for a password" Negative Prompt. You don't need any negative prompt at all unless you want more colorful gens for which you usually add greyscale, monochrome to neg. And 3D just in case you don't want 3Dish look. So to surmise, you only need 3d, monochrome, greyscale, That's it.
Since this Model is trained on AutismMix(well more like lorer merged in it because I'm not Bill Gates), which uses pony diffusion, It follows the same proompting for positive score_9, score_8_up, score_7_up, source_anime, Just use this to start and add your proompt.
No, you don't need quality embedding use them only if you know what you're doing not because your favorite Youthuber or some obscure rentry which the writer boasted on being a proompting God told you to.
No, lorers made on 1.5, 2 does not work on pony.
No. lorers made on XL, Animagine doesn't work on pony. You won't get the characteristics of the concept that it was trained on.
Make use adetailer for face with @Anzhc face segmentation model (it's better than default).
Recommended config
Sampler: Euler a / Euler a Turbo
Sampling Steps: 20-30
CFG Scale: 6-7
Upscaler: 4x-UltraSharp
Denoising strength: 0.2-0.7
VAE: SDXL Vae
Enable Quantization in K samplers