Trained on a handpicked collection of AI generated images (Midjourney, DALL-E) that screams Fantasy and beauty. Originally, I was planning to make something like Pastel-Mix, but I couldn't get good data that doesn't break ??????? like the character is a piece of floating cloth. :pensivePatting:

Note: some images use Controlnet in samples

How to Proompt

• 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 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.

• Make use adetailer for face with @Anzhc face segmentation model (it's better than default).

Recommended config

Sampler: Euler a

Sampling Steps: 20-30

CFG Scale: 6-7

Upscaler: 4xNomos8kDAT, R-ESRGAN 4x+ Anime6B, 4x-UltraSharp

Denoising strength: 0.2-0.7

VAE: Baked in, but you can use any sdxl vae

Enable Quantization in K samplers

Cons:

• The Model sucks at skin colors.