at the moment it is midjourney and DALL•E-2, but a very serious and interesting contender, which is also opensource, is Stable Diffusion. There is also Diso Diffusion, Starry AI and ma y more
Amazing well done!
Wow,thanks…I definitely should dive deeper into this
Tell the Robot what to do and what to think I always say:
Before:
But only have they abide by the 1st Law of Robotics
After:
Looked up some codes on MidJoureny. Use this prompt as a starter if you like
/image prompt <the dream of the eternal one, by Moebius, --ar 16:9 --hd --q 2>
the text in the < >
Change the Text Prompt to whatever you want, the secret is the --ar, --hd, --q settings
ar = Aspect Ratio
hd = High Definition
q = Quality (default is 1) - I’m out of GPU cycles for the month, so I’m looking for ways around the lock out
Please let me know if you have any other ‘cheat’ codes
Here’s an example after 1 Single Upscale from the initial results:
what the fuuuuuu?!
Dude, that’s actually mind blowing
Thanks. I’ve got several more planned. Just need to source the right imagery.
No music tonight, but i am enjoying the rhythm of creating these
Thanks for indulging me ™
You can grab a .pdf of it here as well:
Man, is it your own creations ? I’m impressed. You are really as good in music than in graphic arts.
I didn’t draw this. It’s the MidJourney system. I’m using the renders to tell a story. I wanted to see if i could achieve a multi panel/page narrative with the output from the system. Hope you enjoyed the story
I see this stuff like Samples or Stock imagery. I’m sorting through tons of output to connect them in meaningful ways. Hopefully it works.
Is it all prompt or are you feeding it images from GIR?
Just prompts at this stage, with n# iterations per image set and tweaks to the prompts until i see something i like. It’s like cooking for the first time without a recipe
It’s by no means 1 and done. It’s more like 20-1 to find something i think works and is visually appealing to me.
Very much like sorting through Samples and Stock Imagery.