Art of Simmie and our World

[The Cryptid System]

Hmmm ok I guess I'll save it then. I wasn't planning on using it for monetary gain anyway. I am planning on adding it to my private art collection! I think with your permission it's fine anyway so I guess I am in the clear to save it then. Sorry if this message sounds weird I woke up in a weird state and can't quite figure out what is wrong.

[TurboSimmie]

If this image of me brings you joy then I'm thrilled that you are collecting it! 😁

[The Cryptid System]

Well feel free to be thrilled cause it certainly makes me happy to add it to my collection 😄

[TurboSimmie]

Here's a couples picture of Phil and I! 😁😍

 

 

It doesn't look exactly like Phil of course, but it's not terribly far off either!

[Ashley]

I don't care how you got that, it's beautiful, wow!

 

[Misha] I wish we could draw with that style, it's clean but detailed. ❤

Edited September 27, 2023 by Ashley

[TurboSimmie]

Thanks! 💚

[Ashley]

Okay so now, how did you get that?

[TurboSimmie]

Okay, so, first of all it is indeed AI generated, I don't want to deceive anyone into thinking we're masterful artists or something. 😆

 

First Phil imported a picture of himself and used the image-to-image setting to get a passible AI facsimile of himself, he then paired it with an AI picture of myself, put them together, and used that composite picture as a starting point for a series of images. We picked the best in that series and used it as a starting point for the next series, and we picked the cutest picture in that set and upscaled it! 😁

[Misha]

OMG Simmie please give more details? Which sites, which settings? We want to try this because we have references for all of us and if even for a reference for our family picture that would be insanely helpful! please please please?

[TurboSimmie]

@Misha  @Ashley Of course!!! 😁 I'll explain how we did it!

 

First go to this site: https://tinybots.net/artbot

 

Go to "Create" and put in your prompt; it may take some trial and error to get the results you want with the prompt, but that's part of the fun. If you want to use an image as a base to go "image-to-image" and upload your pic. I recommend cropping and/or resizing your image to a 512x512 square, as the site outputs in that size. (it can do bigger sizes but it may require "kudos" to do so. It'll give you 512x512 squares all day long for free though. You can always upscale the images too.)

 

Some of the settings you might want to play around with such as the steps or guidance, but we use 30 steps and 9 guidance most of the time. The most important thing is which image model you want to use! It's good to experiment with different ones to see which one aesthetically works for you, but the one we used to create the image we posted here was Dreamshaper. Two of our other favorites to use are CamelliaMix2.5D and Pretty2.5D, which both make more drawn/anime style images. You can even set it to use multiple image models. We tend to generate things in groups of 5 or 10 in hopes that at least one of them will be nice. (We've had situations where literally all of them have been nice, and other times they have all been completely awful, and everything in between)

 

You can set the images to be upscaled automatically (which blows them up by a factor of 4) but you can always upscale images after they've been created (which we prefer doing).

 

If you do this please please share some of your creations with us!!! 😁💚