Very Cool AI Image Stuff...
 

So, I have been playing around with the new Stable Diffusion - Very cool but of course you can't produce any "restricted" images - UNTIL NOW!..

If your PC is more state of the ark than state of the art you can use Colabs - Just follow the instructions this guy shows you and you will have all the naked pics of your favorite star in no time - lol....

Don't panic - I managed to get it working so I am sure anyone can!...

https://www.youtube.com/watch?v=x4ZsZtlJcsM

I have a good CPU and an RTX 2080... what computer do you have? You generating AI images locally?

Imagine when AI can produce XXX movie clips...

My old thing can't run it I am afraid - A new one is on the shopping list...

Using Colabs will get me by at the moment.....

I will check it out... gotta finish my site that will make the Tube sites look like model t fords first...

Bro, Stable Diffusion isn't new. It was one of the first AI Image generators available.

It's a new version of it - Just a few days old - Play around here - https://clipdrop.co/stable-diffusion and prepare to be impressed...

No naughty stuff though - That's why you need the Colab workaround or a flash computer.....

The new SDXL model looks really nice

There's a ton of different ways to run Stable Diffusion without needing to buy a computer.

You can add $10 to runpod and play with that thing for hours. You'll get access to very fast GPUs. No need to buy $1000s worth of hardware.

And they are all without filters. Just make sure you use a good model. Can find models here.

https://civitai.com/

Damn - That's me getting no work done today!..

After you run a pod. You'll need to download the model on the pod.

click connect on the pod

go into the jupiter lap port

navigate to this folder workspace/stable-diffusion-webui/models/Stable-diffusion

open the terminal

go to this url find the download url

type in "wget https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE/resolve/main/Realistic_Vision_V5.1.ckpt" in the terminal

go back to the pod click connect and this time go to HTTP Service Port. That's where Stable DIffusion itself is running. On the top left click refresh and you will see the new model.

The vanilla model is kind of trash. But these custom models are awesome.

Thank you:thumbsup...

img2img is alot of fun. Take a picture add AI noise. You get some crazy stuff.

Just installed Control net. I am trying to learn how to use the same model. I hope to make short videos of some story ideas I have.

https://i.imgur.com/ifJy5KG.png
https://i.imgur.com/40CxUwP.png

:1orglaugh:1orglaugh:1orglaugh - WTF? - That is so creepy!..

so this is what you look like before your morning coffee :upsidedow

no matter, still a cutie pie

img2img, Turn the noise level down to get an image closer to the original. I added prompts angry tattoo horror.