#AI Image (and possible video) Model needed for artistic purposes

1 messages · Page 1 of 1 (latest)

unborn glacier
#

Question to the savy folks here... please point me in the right direction:

Im looking to create art for my upcoming novel and all the online AIs so far come with severe limitations, censorhip en masse and worse.

I need an Image AI that

  • lets me create content as uncensored as possible. (I dont create porn, but from sleazy underworld depictions to crimescenes and romance.... and everything in between)
  • It should have reasonable limits for the price and have high res images (2k or more) per default (lots of image AIs seem to throw 1024x1024 images at you, which is definitely to little)
  • Offers some form of creating a persona that I can reproduce consistently in images.

I dont mind paid service and could even do "Offline" models, if they are reasonably easy to set up (Im not a IT specialist 😄 )

smoky lantern
#

then use it to generate images of the specific lora with proper caption/prompt

#

also there are some LLM options to help composing the story or brainstorm the ideas

unborn glacier
#

If I heard that correctly, then a Lora is trained with images, images I dont have and why I need an AI to create them in the first place...

For lots of things I currently use Gemini, but creating consistent images isnt its strong suit.
What ways / sites / apps can you recommend to train a lora?
I think they mostly use stable diffusion, correct?

(So far I only used the model trainer of getimg.ai. I had decent success with it, sadly only one model could use that and it only output 1k res images)

smoky lantern
#

if you find them somewhere in the internet, grab them

unborn glacier
#

Im would need models for the characters from my novel to reproduce them consistently in my book. Im not trying to reproduce real life people.

Unless I misunderstood something?

smoky lantern