Stable Diffusion, NovelAI, Machine Learning Art - AI art generation discussion and image dump

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
notorietus said:
OpenAI introduced SORA (Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.) today on their Twitter thread - archive. You can visit it here.
View attachment 5725089
Here is an example. The prompt is: "Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes."

View attachment 5725088
Click to expand...
Oh cool! Time for more man-made horrors beyond my comprehension!
 
  • Feels
  • Like
Reactions: Bungdit Din, iSaidShhh, Edict of Expulsion and 4 others
Badgers - Mughal 1.jpgBadgers - Mughal 2.jpgBadgers - Mughal 3.jpgMonkey - Sad 1.jpgMonkey - Sad 2.jpgMonkey - Sad 3.jpgOphanim 1.jpgOphanim 2.jpgSnail.jpgSmug Jug 2.jpgSmug Jug 3.jpgDormouse Tile.jpgTerracitia 1.jpgTerracitia 2.jpgBadgers - Feudal.jpg
Smug Jug 2.pngSmug Jug 3.pngSnail.png
smug.png
I've found it much more fun than I should to make transparent reaction images in various historical styles
 
  • Like
  • Winner
Reactions: Bungdit Din, PureP, AssignedEva and 22 others
notorietus said:
OpenAI introduced SORA (Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.) today on their Twitter thread - archive. You can visit it here.
View attachment 5725089
Here is an example. The prompt is: "Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes."

View attachment 5725088
Click to expand...

I love the Anti-Christ.

Not actually sarcasm, I think this is great.
 
  • Agree
  • Like
  • Thunk-Provoking
Reactions: Ether Being, woowie queen, TheGuntinator and 3 others
Breadbassket said:
Just focus on the subtle issues with videos to identity if they are AI made or not (height issues, distortions, strange angling, clipping), the bad part is this stuff will most likely be ironed out as time goes on which will make everything even harder to recognize.
Click to expand...
You know it'll only get better and better, right?
The genie is out, let's see what we do with it.
 
  • Optimistic
  • Like
Reactions: TheGuntinator and Sinner's Sandwich
notorietus said:
Here is an example. The prompt is: "Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes."
Click to expand...
When this gets open-sourced we’re all fucked. There is going to be so much weird porn the coomer brains will finally fry permanently
 
  • Like
  • Agree
  • Optimistic
Reactions: imaginary friend, woowie queen, TheGuntinator and 6 others
Super Dilator Duel 3D said:
When this gets open-sourced we’re all fucked. There is going to be so much weird porn the coomer brains will finally fry permanently
Click to expand...
I've always figured that's going to go the same way the inevitable sex bots will -- it'll just finally permanently remove the coomers from society entirely where they can't do any real harm. When you can just tell your computer to produce high-quality porn of any kind to your exact specifications, you won't need to piss away money on OnlyFans (so it'll help kill that parasocial relationship crap) or paid porn anymore. Sex bots will be for the more "adventurous" coomers who want to go beyond watching a screen.

Either way, it ultimately saves money and keeps them indoors and out of harm's way.
 
  • Thunk-Provoking
  • Agree
  • Optimistic
Reactions: Irrational Exuberance, AFAB, TheGuntinator and 7 others
Slav Power said:
1) Can I run this locally
2) How much VRAM and processing power is needed to run it compared to SDXL and SD1.5
3) Is it the same level of open to training to generate anything as SDXL and SD1.5
Click to expand...
Yes, you can run it at home already now but I'd give it a few days. On a A6000 the generation takes a few secs (not instant but close to) per picture at around 12 GB of VRAM. It also has such sights to show you.

cyb1.jpgcyb2.jpgcyb3.jpgbugs.jpg

It's a lot more coherent than when I tried XL last which admittedly was a while ago. It's *almost* like it understands the prompts actually, instead of just throwing random things at you that are somewhat related to the words you wrote. It also is a lot less trending towards *accidential* body horror, I even saw the one or other plausible keyboard. According to stable this is a very early version of things to come.
 
  • Like
  • Informative
  • Winner
Reactions: Vecr, The Mass Shooter Ron Soye, TheGuntinator and 7 others
notorietus said:
Here is an example. The prompt is: "Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes."

View attachment 5725088
Click to expand...
A lot of stock footage repositories for youtube videos are going to dissapear very soon
 
  • Thunk-Provoking
  • Agree
  • Like
Reactions: Jimmy Hopkins, Surfherder, woowie queen and 3 others
AmpleApricots said:
Yes, you can run it at home already now but I'd give it a few days. On a A6000 the generation takes a few secs (not instant but close to) per picture at around 12 GB of VRAM. It also has such sights to show you.

View attachment 5725355View attachment 5725356View attachment 5725358View attachment 5725359

It's a lot more coherent than when I tried XL last which admittedly was a while ago. It's *almost* like it understands the prompts actually, instead of just throwing random things at you that are somewhat related to the words you wrote. It also is a lot less trending towards *accidential* body horror, I even saw the one or other plausible keyboard. According to stable this is a very early version of things to come.
Click to expand...
My main gripe is can it offer me better results than SD1.5 with the same generation time as SD1.5 on my poor shitty 1060, or am I perpetually stuck in the SD1.5 limbo until I get a modern GPU around never.

If I'm reading this correctly, it's actually less demanding than SD1.5, so in theory I should be able to generate 1024x1024 images on my 1060 with better quality at lower generation times, making this a really fucking big deal. Am I right?
 
  • Feels
  • Like
Reactions: Vecr, Ether Being, Edict of Expulsion and 1 other person
I tend to believe in reality it will still have artifacts even on future models.
I would like to run something like this but my PC specs suck.

In the worst case, that it is impossible to detect artifacts, audio-visual content will not be allowed in courts and we will go back 100 years like @9gfuwegw9j9 said.
 
  • Agree
Reactions: Hyperbolic Steroids and Klaptrap
Guess I'll be waiting for another month before I consider Stable Cascade. Is it compatible with SD 1.5 LoRAs?
 
You must log in or register to reply here.
Back