theUnlikely

joined 2 years ago
MODERATOR OF
[–] [email protected] 9 points 1 year ago (14 children)

I had a feeling that Stable Diffusion wouldn't be able to generate what I was looking for, and I was right. So I turned to Dalle, which gave me some okay, but bland images. Then I took ideas from those and went to MidJourney where I got some beautiful images, but just not quite right. But I got some more stylizing ideas from those. Back to Dalle and got a good, but still bland image. Now it's SD to the rescue!

I started with inpainting the the surrounding cityscape bit by bit using one of my favorite realistic models Photon. Then with the Unstable Diffusers Nihilmania SDXL model, I proceeded to fix up the crack in the street. First the outline of the cracks, and then the inside with the Pixel Art XL lora.

Original Dalle prompt:
realistic digital art, a deserted city street at dusk, shimmering with the soft glow of street lamps, parts of the asphalt crack wide open, subtly revealing bright, pulsating binary code beneath, widescreen aspect ratio

[–] [email protected] 1 points 1 year ago

Thanks a lot for making this! I'm starting to have some luck with the Pony model finally. It's crazy how many resources that model has already.

[–] [email protected] 1 points 1 year ago

This account on civitai has many many good examples, but all SD1.5 https://civitai.com/user/TxcTrtl/images?sort=Most+Reactions

[–] [email protected] 1 points 1 year ago (2 children)

Sure! The main image of this post is one example.
And here are a few more (all from SD1.5 models):

spoiler

[–] [email protected] 1 points 1 year ago (5 children)

A comfyui json would be much appreciated!

[–] [email protected] 1 points 1 year ago (7 children)

I usually go with the full models, but for some reason GrogMix only has turbo and can nail the look I'm after.

I keep seeing positive sentiments about Pony everywhere I look, so I'll keep trying with that one. Right now, I'm having trouble with any character that isn't very close. Anything slight back from super closeup portrait loses a ton of detail, even with hires fix and multiple rounds of upscaling.

[–] [email protected] 1 points 1 year ago

Definitely! That's how I was able to find the ones I listed in another comment. Without this, I'd be doomed 😆

[–] [email protected] 1 points 1 year ago

Yeah it seems like loras are necessary. For that image I used the concept art twilight lora.

[–] [email protected] 3 points 1 year ago (2 children)

I did that with RealCartoon-XL, and soooometimes it gave good results for the style, but inconsistent. I guess I just got too used to SD1.5 models specializing in a certain aesthetic.

[–] [email protected] 1 points 1 year ago (2 children)

Thanks for the reply, but yeah, I've definitely tried that. In ComfyUI for this model, setting it to -2 doesn't actually do anything compared to no node for that at all. Setting it to -1 of course ruins it though.

It seems really temperamental with the styles. It seems to vastly change just by changing a few words that aren't related to style at all.

This is the best I've gotten with it after hours if fiddling. Most of the image is good, but the face and hands never are.

[–] [email protected] 2 points 1 year ago (4 children)

That's actually a good example of the overly airbrushed look I was trying to describe. So many models have output like that.

As for GrogMixTURBO, I found the keywords "thick lines, flat shading" from one of the creator's replies to a comment on the model page, and these magically gave me what I wanted with that model.

[–] [email protected] 1 points 1 year ago* (last edited 1 year ago) (4 children)

Ones I've found so far are:

  • GrogMix TURBO
  • _CHEYENNE_
  • RealCartoon-XL (sometimes?)
  • PixelPaint
  • PixelWaveTurbo

I've seen good images using Pony XL, but my attempts with it have so far produced messy slop so I'm not sure what I'm doing wrong there.

view more: ‹ prev next ›