I agree about missing the 'old model'. The thing I needed - and liked about it - was consistent facial features. The new one is extremely difficult to get that. I need the consistency as I use images for illustrations in stories that have characters. With the randomness that happens now it is unusable for my needs. I enjoyed it while it was fitting that aspect of what I was doing.
The new model is hot garbage. It will not reproduce any of the hundreds if not thousands of faces it easily produced before. Now it had 4 male and 4 female faces varied by skin tone. The background detail is the slightest bit better, but not worth trashing what it was. I spent over a year on hundreds of prompts and no not a single produces anything remotely like before. Its 1 step forward and 32 steps backwards. Sad change indeed.
I think dev made up his mind on having Flux.1 be permanent for all of Perchance AI image generators.
The people who adore the new model have won over the Developer of Perchance.
Pretty ridiculous and disingenuous thing to say. The update has been planned and mentioned in this forum for a while. This was not a "test" to see which "side" would like better so that only then the dev would make their final decision.
The dev likely had made up their mind way, way before the model was implemented, otherwise they wouldn't have implemented a new model, obviously.
It's just a thing to point out to people that still want the old model that is pretty much pointless to ask for the old model back and stuff.
Personally speaking It's become useless to me..whether that will change in 6 months,who knows?..just feels like it's gone backwards..what happens if we all get used to this version then he decides to "upgrade" it again & we're back to square one?..sometimes new isn't always better.
The Flux model is objectively better at everything except speed and celebrity recognition (which is technically illegal anyway). People just need to familiarize themselves with the new style of prompting. Just write as if you were describing the image to someone with a mental deficiency and you'll get some great images.
Yes but it's not just the fact that it no longer recognises "celebs",it's the fact that some of the old styles are nothing like they were - tintin comic,franco/belgian comic,50s enamel sign,vintage comic,vintage pulp art etc etc are now Nothing like they were or are anywhere near as good..how is this progress?..surely these styles should've stayed as they were even if there is some sort of new engine running this site.
I didn't expect that there would be bug to tag level
We need a comeback of the old model which was SD1.5 not SDXL
This is a Lemmy Community for perchance.org, a platform for sharing and creating random text generators.
Feel free to ask for help, share your generators, and start friendly discussions at your leisure :)
This community is mainly for discussions between those who are building generators. For discussions about using generators, especially the popular AI ones, the community-led Casual Perchance forum is likely a more appropriate venue.
See this post for the Complete Guide to Posting Here on the Community!
1. Please follow the Lemmy.World instance rules.
2. Be kind and friendly.
3. Be thankful to those who try to help you.
4. Only post about stuff related to perchance.
5. Refrain from requesting Prompts for the AI Tools.
text-to-image-plugin and ai-text-plugin) e.g. "What is the good prompt for X?", "How to achieve X with Y generator?"6. Search through the Community Before Posting.