A completely unnecessary way to generate new prompts from existing images. #StableDiffusion2 #ChatGPT #aiart
Last one. Butcher Billy's examples are pretty even for both but SD2's Christopher Balaskas is very flat and boring. Don't think it got his style very well. #StableDiffusion2
One trend I've noticed is that SD2 feels more incoherent and noisy compared to 1.5, yet also lacking in interesting details. I guess it could work for certain artists. Wayne Barlowe creates interesting creatures, but also more random artifacts in SD2. #StableDiffusion2
Big thanks to @KyrickYoung and @proximasan for putting together the #StableDiffusion2 artist list. Here's some comparisons I made between SD2 and 1.5 using a basic prompt. Tried to pick out varying styles. (Long-ish thread)
Left side: SD 1.5 (512x512)
Right side: SD 2.0 (768x768)
Had to completely rework my prompt for this comparison. So far, Stable Diffusion 2.0 has nice realism but less imaginative? The lack of certain artists is definitely noticeable but I'm still experimenting.
Top row: 1.5
Bottom row: 2.0
#StableDiffusion #AIArtistCommunity #aiart
And after some experimental merging with other models, I'm able to incorporate more art styles and backgrounds on my character. #StableDiffusion
Stable Diffusion 1.5 is great but I think the real star is the new inpainting model. Both inpainting and outpainting is so much easier now. #StableDiffusion #AIArt #StableDiffusionArt
Just trained two Mirror's Edge hypernetworks. One is for rooms and environments (left image), while the other is for more abstract art (right image). But feel free to experiment.
Normal - https://t.co/oSiJmb8iqc
Abstract - https://t.co/rFC4twmdIL
#StableDiffusion #AIArt
@SmirkingDesigns @NiphionArt @Jayexe8 Not sure about Prince Diamond but I got a pretty close Sesshoumaru. Doesn't quite get the face markings right but still. Prompt is in the image alt text.