[-] beautifulsoup@lemmy.world 3 points 1 month ago

They're training the model at a very basic level so hopefully they'll add celeb LORAs in future

[-] beautifulsoup@lemmy.world 2 points 1 month ago

There's no law

[-] beautifulsoup@lemmy.world 3 points 2 months ago

Literal 0 improvements in the last 2 weeks

[-] beautifulsoup@lemmy.world 2 points 2 months ago

We need a comeback of the old model which was SD1.5 not SDXL

[-] beautifulsoup@lemmy.world 5 points 2 months ago

Yeah, but this was not some bugfix or minor change, was it? They replaced the 1.5 years old core model itself; don't you think they should had made any announcements, like where they're showing NOW 'Notes on new model', it could have been 'Notes on upcoming new model'

[-] beautifulsoup@lemmy.world 2 points 2 months ago

Try using different browser

[-] beautifulsoup@lemmy.world 2 points 2 months ago* (last edited 2 months ago)

Ok then state this model's Pro aspects and it has to be more than 4. I know one of them would include furry and cartoons anyways

[-] beautifulsoup@lemmy.world 7 points 2 months ago

The way the devs blatantly replaced the old model overnight, like it was worthless or something, shows how little they care about their own community

[-] beautifulsoup@lemmy.world 2 points 2 months ago

Any news on this?

[-] beautifulsoup@lemmy.world 4 points 2 months ago

Perchance has downgraded itself

[-] beautifulsoup@lemmy.world 5 points 2 months ago

It's a stretched solution and does not count as a positive side of this model either. Actually, you'd know the issue in the first place what I meant by not creating 'consistent faces' if you had been working with realistic faces for a year or so, which I doubt

[-] beautifulsoup@lemmy.world 2 points 2 months ago

I wasn't talking about anime or furry faces bruh

7
submitted 2 months ago* (last edited 2 months ago) by beautifulsoup@lemmy.world to c/perchance@lemmy.world

Just to be clear right off the bat, I DESPISE this change.

Pros:

  1. Generated images can be very realistic and of high quality.
  2. New image size is available.

Cons:

  1. Generates different people with different facial features every time, there's no consistency, no matter how many different prompts or entire sentences were provided.
  2. Even after trying to avoid weird, deformed images one'll get a few straight-up UGLY AS SIN or absurd images in between a number of generations. I swear I'd never get anything even close to being "ugly" with the previous model, unless I prompted something like 'Cursed photo'.
  3. The model can't draw any resemblance to celeb faces anymore, even if one tries, it generates a random face.
  4. The amount of time it takes to generate now is around 3x longer, sometimes it just fails and retries.

Lastly, the previous model, which had been in use for over a year, was replaced overnight, like literally. All the prompts that everyone had created over so long and with care, suddenly became meaningless. I had to create an account just to check if it's only my pain that I've to bear.

view more: next ›

beautifulsoup

joined 2 months ago