🔞 NSFW 🔞 Stable Diffusion / AI art thread "Alien technology that allows for superpowers" edition - featuring a helpful guide to help you install SD

[r4v3r]

robo

 

[Sen]

.

 

[Xeraser]

marisa sketches (not mine)

 

[Xeraser]

i fully support our robot overlords

 

[Xeraser]

got my wife in a track suit
god bless

 

[succuri]

There ain't any content in this thread that is NSFW!! Please consider changing the title. Thanks

 

[Xeraser]

In case you've all missed them, there have been MASSIVE advancements in terms of training tech. Both hypernetworks and embeds have been superseded by LoRAs, extremely fast and efficient mini-models that take a fraction of the time and resources to create when compared to dreambooths (fine-tuned models)

 

[nikey]

In case you've all missed them, there have been MASSIVE advancements in terms of training tech. Both hypernetworks and embeds have been superseded by LoRAs, extremely fast and efficient mini-models that take a fraction of the time and resources to create when compared to dreambooths (fine-tuned models)

Click to shrink...

couldn't find anything about LoRAs, please give me a link

 

[Xeraser]

couldn't find anything about LoRAs, please give me a link

Click to shrink...

GitHub - cloneofsimo/lora: Using Low-rank adaptation to quickly fine-tune diffusion models.

Using Low-rank adaptation to quickly fine-tune diffusion models. - GitHub - cloneofsimo/lora: Using Low-rank adaptation to quickly fine-tune diffusion models.

github.com

Potentially nsfw: https://rentry.org/lora_train

 

[Xeraser]

i should make a new guide tbh

 

[D00GL]

anybody know a good way to train my model to put welding masks on girls heads without degrading the style? last image was with dreambooth at idk what settings. i have about 50 training images with prompts.

 

[Xeraser]

I added Jeremy Clarkson to my prompt and it rendered her more on-model.

 

[Xeraser]

one click stable diffusion on ios/ipad os & m1/2 macbooks https://drawthings.ai

 

[Capybara Connoisseur]

This website has a bunch of models.

Civitai | Stable Diffusion models, embeddings, LoRAs and more

Civitai is a platform for Stable Diffusion AI Art models. Browse a collection of thousands of models from a growing number of creators. Join an engaged community in reviewing models and sharing images with prompts to get you started.

civitai.com

 

[lizzyrascal]

my favorite interface for SD and other models is ComfyUI <3

although some models like to break, I struggled to get PonyDiffusionXL to work properly, but when I did; the result was worth it

The YiffyMix Model + FurryCore LoRA makes extremely good yiff (all models i mentioned can be found on civitAI)

if anyone wants a comfyUI workflow for Pony or the YiffyMix lemme know

 

[PeavyFox]

AI FOX

 

[Xeraser]

my favorite interface for SD and other models is ComfyUI <3

although some models like to break, I struggled to get PonyDiffusionXL to work properly, but when I did; the result was worth it

The YiffyMix Model + FurryCore LoRA makes extremely good yiff (all models i mentioned can be found on civitAI)

if anyone wants a comfyUI workflow for Pony or the YiffyMix lemme know

Click to shrink...

Sorry, cumragUI is ***OUTDATED***
Forge is infinitely better.

 

[lizzyrascal]

Sorry, cumragUI is ***OUTDATED***
Forge is infinitely better.

Click to shrink...

Yknow after seeing this I actually checked out Forge and its actually pretty badass :D runs pretty fast and is straightforward, just wanted to say thanks!

What I've done is just moved my models over to their own folder and set the paths of both UIs accordingly so i can use both Forge and ComfyUI

 

[Kevin Logan]

 

[dainiswilliams]

i fully support our robot overlords

Click to shrink...

This art is made by - https://www.nsfwartgenerator.ai/