Type | |
Stats | 3,003 |
Reviews | |
Published | Apr 12, 2024 |
Base Model | |
Hash | AutoV2 4AD54EB171 |
As this is a merge based off PonDiffusion v6 I HIGHLY recommend you read the description of that model for information regarding prompting, tags, and baked in characters.
The baked-in artist styles for PD should work, but as this is effectively a style merge, they will defeat the purpose of this model.
Generation settings I use:
Sampler - Euler a / Euler a SGMUniform
Sampling Steps - 25
I upscale to 4x using this process and the upscaler below:
Dimensions - 832x1216, 1216x832, or 1024x1024
CFG Scale - 7 or higher
One or more passes of ADetailer
Some combination of the following for faces and eyes
I default to the following positive and negative prompt. I DON'T USE ANY SOURCE TAGS (source_anime, source_furry etc) :
score_9, score_8_up, score_7_up, BREAK {prompt}
No negative prompt required
Prompt for what you don't want to see as needed (patreon, teeth, simple background etc.)
See sample images for examples.
A personal stylized merge I've been tweaking on and off for a few weeks. I wanted to have a checkpoint that simulated the style of a few artists I really enjoy without having to resort to using multiple LoRAs or schizo prompting. Specifically, I wanted a high amount of consistency without any negative prompting. I ended up being very pleased with the thicker outlines and slightly more washed out colors (hence the name), so I figured I would share it. This is still pretty experimental and probably not worth 6.5gb of your hard drive, but I'm too lost in the sauce to care.
As with most of my models, this is intended primarily to produce female characters and leans towards NSFW content.
I am still no expert when it comes to SDXL (and SD in general) so I would recommend experimenting with these settings to find what suits you.
All credit should go to the original model authors, I'm just mashing stuff together. Please check them out: