Type | |
Stats | 402 |
Reviews | (103) |
Published | Mar 12, 2024 |
Base Model | |
Hash | AutoV2 CE410F56A7 |
As this is a merge based off PonyDiffusion v6 I HIGHLY recommend you read the description of that model for information regarding prompting, tags, and baked in characters.
With the huge influx of artist and style LoRAs accompanying the advent of PonyDiffusion v6, I wanted to make a blend merged into AutismMix_pony. Heavily stylized to my personal taste, may not be your cup of tea (although I don't think it's difficult to figure out the influence). Still very much experimental, and I cannot yet guarantee the consistency of outputs or compatibility with all LoRAs. Hands in particular are still pretty hit or miss. Models I used are listed at the bottom.
As with most of my mixes, this is intended to produce female characters and leans towards NSFW.
V2:
Remerged from the bottom-up using modified weights. Less stylistically intense (especially around the faces and eyes), more inline with what I had originally intended for this model, but still not quite there. Reduced the need to use specific prompts to achieve the desired look. Hands are still hit or miss, and there's a persistent watermark that appears ~50% of the time: some inpainting might be required. Hoping to rectify this in a future update.
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 - DPM++ 2s a/DPM++ 2s a Karras/DPM++ 2M Karras
Sampling Steps - 25
I upscale to 4x using this process and the upscaler below:
Dimensions - 832x1216, 1216x832, or 1024x1024
CFG Scale - 8 or higher
One or more passes of ADetailer
Some combination of the following for faces and eyes
I tend to use the following positive and negative prompt. :
score_9, score_8_up, score_7_up, BREAK {prompt}
no negative prompt required
watermark, patreon username, artist name, signature will help with watermarks generating
simple background in the negative gives very vibrant background
See sample images for examples.
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: