Crumb
c/perchance · by OrbitalNuke@lemmy.world · 2y

What model is used for ai-text-to-image-generator?

https://perchance.org/ai-text-to-image-generator

Could we self-host it?

6 points · 8 comments · view on lemmy.world

8 Comments

VioneT@lemmy.world · 3 pts · 2y

Please refer to this post about the model.

Golfnbrew@lemmy.world · 2 pts · 2y (4 replies)

Interesting. I asked for a snowy forest, no people, and I ONLY get people. Standing in a snowy forest, yes, but not what I asked for.

eatham@aussie.zone · 1 pts · 2y (1 reply)

try not mentioning people

Golfnbrew@lemmy.world · 1 pts · 2y

Yeah, did that first. "snowy forest"...

surewhynotlem@lemmy.world · 0 pts · 2y (1 reply)

You need to put the people as a negative prompt. Not sure how to do that in this UI. but it's possible in automatic 1111

Golfnbrew@lemmy.world · 1 pts · 2y

Yep, did that too...

catbat@lemmy.world · 1 pts · 2y

more info here: https://lemmy.world/post/9542946

synapse1278@lemmy.world · 1 pts · 2y

For self hosting, you can look at Automatic1111, it's not the same web-interface though, A1111 is fair bit more complex.