SDXL v1.0 is out and initial testing has shown it to be much higher quality than SD v1.5.
After comprehensive testing in a variety of different art styles, I've concluded that it is much much better than 1.5 by a staggering margin.
I made sure to test across a range:
- Natural Language Processing: Recognition of characters and art styles
- Art styles: Greater understanding of nuanced styles
- Ease of use: Extremely flexible even when posed challenging prompts
- Functionality: Handles generic and specific(highly detailed) prompting well
and it's been simply amazing.
Recommended Settings
Some personal recommendations after extensive testing:
Resolution
SDXL v1.0 was trained on larger images so stick with dimensions bigger or equivalent to 1024 x 1024. The results are inconsistent when trying to go smaller than this.
I've found that the default 1024 x 1024 works pretty well.
Samplers
I highly recommend the DP++ samplers, they are a relatively newer family of samplers and have good results. If you are unsure, just start with .
- : gives good clarity but faster
- : gives strong quality but is a bit slower
How to Run
Since the models for SDXL v1.0 are more complex, running on M1/M2 macs currently isn't enjoyable. I recommend using a free service for image generation. Check out my guide on how to get started here.
- [TLDR]: Sign up for an account on Tensor Art (they have a generous free image generation that refreshes daily.)
1. Bonsai World
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
2. Yellow Muted
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
3. Pixel Art
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
3.1 Cozy Cafe Pixel Art
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
4. Isometric
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
5. Fantasy World
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
6. Wes Anderson
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
7. Cyberpunk
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
8. Origami
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
9. Low Poly
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
10. Clay
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
11. Anime
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
11.1 Tokyo Bridge Overlooking Water
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
12. Cinematic
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
13. Photography
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
14. Comic Book
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
15. Analog Film
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
16. Polaroid
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
17. Fantasy Art
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
18. Surrealism
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
19. Takashi Murakami
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
20. Vaporwave
Stable Diffusion v1.5
SDXL v1.0
Prompt:
Loading...Negative:
Loading...Seed:
Sampling Steps:
Sampling Method:
About @workingforlatte
Hey there, I'm @workingforlatte, an engineer and solopreneur on a mission to make AI accessible and affordable while indulging in my love for lattes. One day, I hope to treat myself to a luxurious mcol latte, and who knows, maybe even upgrade to an hcol latte (a lofty dream, I know)!
With a deep passion for technology and AI, I bring my engineering expertise to the forefront as I explore the ever-evolving AI landscape. I'm thrilled to be part of this AI blog, sharing my unique insights and unraveling the mysteries of artificial intelligence.
Join me on this exciting journey as we navigate the intricate world of AI together. From tackling complex technical challenges to discovering the endless possibilities of AI, I'm here to inspire and empower fellow tech enthusiasts like you.
Together, let's embark on this adventure of uncovering the wonders of AI and embracing the incredible ways it will shape our future. Get ready to dive into the AI realm alongside me, and let's make AI knowledge accessible and enjoyable for everyone. Cheers to a future filled with AI breakthroughs and, of course, plenty of delicious lattes!