With thanks to Beachy, I am going to star tmy own thread of pointless images that I've created using this site...

https://you.com/search?q=who+are+you&tbm=youchat

...under the 'imagine' tab and using Stable Diffusion 2.1

So, here we go...

"Yeti"



"Bigfoot in California"

"Andy Warhol's Jesus"




"Andy Warhol as Jesus"

You've gotten some really nice face images. I think since most of my reference photos have been coming from
the 1930s, the AI just hasn't had the opportunity to look at anything high-resolution.

That might be the case, yes.

"Jimi Hendrix as Jesus"

And I guess I've been using Stable Diffuse 1.5 more than 2.1 because my Internet is awfully slow. 1.5 renders faster.

"Farrah Fawcett as Supergirl"



"Gorilla in a Tuxedo"



"Electric Guitarist with long hair"

"Superhuman"



"Apeman in a spacesuit"

I tried this out and got one to complete, although it was a disappointment. Probably I should have described it better, dunno. Then it made me sign up (which was fine), but it's taking ages to generate an image. Been running for several minutes now. Do these take you that long?

Less than one minute, normally.

"Chimpanzee as Superman"



"Robot Monkeys"

I've found that if I try to run a second one, it takes way too long.
So I normally run one, kill the browser connection, then start up a new session.
I don't know why that helps, but a fresh session runs much faster, even faster if I haven't been
to the site for awhile. It may have some built-in queuing software to give more user sessions access.

Another thought is that I know the AI in the chat version learns from your sequence of requests within
a session, so maybe it's trying to render your next image while still considering the history of your previous
image request. Starting a new session each time speeds it up.

This is not mine but it's AI Generated and I like it and I thought others might also like it.

I got it to work more reliably on my phone. The results were... interesting. I used "Marilyn Monroe meets Spider-Man" and generated a few of them. Didn't save any of them though; nothing was really worth seeing.

I beg to differ...I'm doing it right now...

...and these were the results...

All much better than the results I got, I assure you. Most were in comic panel format, with an odd mix of drawn/colored comic and "live" Marilyn.

Marilyn Monroe and Superman got me this one, which is pretty cool.

The Anime model got me this:

Spider-Man, Marilyn Monroe, Mary Jane Watson Love Triangle


Marilyn's upper body doesn't quite seem attached, but otherwise, not bad at all.

A clean session of SD 1.5. All I typed in was "IMWAN":



Looks peaceful and irrelevant.

Not. A. Fan.

So this is "Marilyn Monroe meets Spider-Man" using the anime model (which I'd never used before). I decided to try a few using the Anime setting just for fun.



This is "The Patterson Gimlin Footage"



"The Who 1974"



Now, the same three images but done with Stable Diffusion 2.1

"The Who 1974"




"The Patterson Gimlin Footage"




"Marilyn Monroe meets Spider-Man"



It's intriguing - the way you can write the exact same command and get such completely different images every time is interesting, but the fact that it blends photos with cartoons is interesting to me. I also like the way it interprets written words. This is fun.

In the options, you can enter a Seed number. I haven't tried it yet, but I think if you use the same Seed and then
enter the same text, you'd get the same images.

But possibly you can enter Seed, Same Text, but alter one of the other settings to fine tune an image.