[Stable Diffusion] Prompt Sharing and Learning Thread

[StepSisWasNeverStuck]

StepSisWasNeverStuck

Btw. I'm using my preference in girls instead of yours because I find them easier on the eye to work with. You can regen and do everything with your choices haha.

Okay, here's the continued workflow.
I took my version 1 top left girl that I pasted last time, and I've basically thrown her into photoshop to get the last stage, regenerated into auto1111 with img2img, then back and forthed with multiple gens, varying CFG scale and lower denoise.

You can see that I changed her outfit by just painting in a style I thought was less like a bikini and more like a corset-dress. I also added wristwear as you can see. With each re-submission into img2img I changed up the prompts accordingly.

Higher CFG = more dreaming.
Lower denoise = closer to original image in the img2img

Add background from wherever you want (I chose a cathedral pic), layered underneath the girl within photoshop.

Rough illustrated workflow is:

View attachment 2537327
Start from bikini. Throw into PS. Scribble in straps, wrist stuff, whatever. Img2img. Find one you like.
Throw into PS. Scribble. Img2img.

Find background
View attachment 2537335

Pull person from old img2img.

View attachment 2537336

Add her into background

View attachment 2537339

Find flames from somewhere. add in.
Img2img to blend.

View attachment 2537341

View attachment 2537340

Add photo filter of choice (i just winged it for fast results).

Continue to tweak.
Forgot to add - I changed to Neverendingdream at some point. Up to you.
Resize with latent upscale if you want to change the image significantly

"I feel like we have reached our goal"

miaouxtoo, I mean the amount of efforts you are putting in creating these amazing images for us to learn from is just out of this world. I don't think i can thank you enough for helping me so much.

And the reason i was trying to copy his style is because there are many people whom I follow who create very perfect and similar looking artwork to these (some examples from twitter)


but i had no clue what they were doing in SD to create something like that. And I was so obsessed with these artwork that i want to create them by myself, so i asked you guys for help and luckily miaouxtoo helped.
I feel like i will be able to create images like these after i do some experiment and apply the concepts that you taught me.

 

[miaouxtoo]

"I feel like we have reached our goal"

miaouxtoo, I mean the amount of efforts you are putting in creating these amazing images for us to learn from is just out of this world. I don't think i can thank you enough for helping me so much.

And the reason i was trying to copy his style is because there are many people whom I follow who create very perfect and similar looking artwork to these (some examples from twitter)
View attachment 2538009 View attachment 2538012 View attachment 2538013 View attachment 2538014 View attachment 2538015 View attachment 2538016

but i had no clue what they were doing in SD to create something like that. And I was so obsessed with these artwork that i want to create them by myself, so i asked you guys for help and luckily miaouxtoo helped.
I feel like i will be able to create images like these after i do some experiment and apply the concepts that you taught me.

Nice! Always glad to help satisfy curiosity in learners. I think as tools get more advanced, the styles of art will improve. Right now everyone is just figuring out how to generate static images ... but movies and actual HQ SD porn (lol) is coming no doubt.

The chun-li in your list of images is most likely a Lora + controlnet mix. The others are img2img & generations with models.
A lot of the time it's the model that makes the difference - aided by loras.

 

[miaouxtoo]

Sometimes you get the girl you want.





And sometimes you get the girl that's sniffed a bit too much glue...

 

[rayminator]

computer spec:
MB.ASUS TUF GAMING B550-PLUS WIFI II
CPU.AMD Ryzen 5 5600G
DDR4.CORSAIR Vengeance LPX 64GB (2 x 32GB)
GPU.MSI Ventus GeForce RTX 3060 12GB GDDR6 PCI Express 4.0
AI timesteps:167 if that matters

can someone try this negative prompts for me cause when I use it and when I don't it come out like this

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

when using it


when I'm not using it

 

[Mr-Fox]

computer spec:
MB.ASUS TUF GAMING B550-PLUS WIFI II
CPU.AMD Ryzen 5 5600G
DDR4.CORSAIR Vengeance LPX 64GB (2 x 32GB)
GPU.MSI Ventus GeForce RTX 3060 12GB GDDR6 PCI Express 4.0

can someone try this negative prompts for me cause when I use it and when I don't it come out like this

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

when using it
View attachment 2539100

when I'm not using it
View attachment 2539101

It looks like you have many words that is repeated. Start by removing all repeats and see what this gets you. I can't do any tests unfortunately because I'm training a Lora atm. If the issues persist I might be able to do some tests tomorrow.

 

[rayminator]

It looks like you have many words that is repeated. Start by removing all repeats and see what this gets you. I can't do any tests unfortunately because I'm training a Lora atm. If the issues persist I might be able to do some tests tomorrow.

I have changed it a little bit like adding : after bad/deformed/double hope the ai understands here a example

is there any extensions that easily combines all this into a tag or something like that?

or is there another way to combine bad/deformed/double

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

 

[Mr-Fox]

I have changed it a little bit like adding : after bad/deformed/double hope the ai understands here a example

is there any extensions that easily combines all this into a tag or something like that?

or is there another way to combine bad/deformed/double

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

View attachment 2539343 View attachment 2539341

Here's a prompt guide I think that you would benefit from reading it. So would most tbh. Probably even myself..

You must be registered to see the links

 

[Mr-Fox]

I have changed it a little bit like adding : after bad/deformed/double hope the ai understands here a example

is there any extensions that easily combines all this into a tag or something like that?

or is there another way to combine bad/deformed/double

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

View attachment 2539343 View attachment 2539341

I don't know of an extension but you can save a prompt as a style in webui. Perhaps there are extensions with more functions but I'm not aware of one.

 

[Jimwalrus]

I don't know of an extension but you can save a prompt as a style in webui. Perhaps there are extensions with more functions but I'm not aware of one.

There are a couple of negative embeddings on Civitai. My favourite is

You must be registered to see the links

 

[rayminator]

Here's a prompt guide I think that you would benefit from reading it. So would most tbh. Probably even myself..

You must be registered to see the links

thanks for the link I will look at it

it's just fix one deformed part then another part of the starts to deformed

I don't know of an extension but you can save a prompt as a style in webui. Perhaps there are extensions with more functions but I'm not aware of one.

I'll keep an eye out for one instead of having something like this

 deformed: ass,butt, glans,clothing,arms,legs,feet,breasts,penis,neck,head,wrist,waist,stomach,nose,ears,eyes,hair,fingers,toes,ankles

you just need to put some like <Tag:deformed.01>

it's just a idea don't know if it possible

 

[Elefy]

Aunt on the river (amateur photo)

 

[Mr-Fox]

View attachment 2540782

Aunt on the river (amateur photo)

Freekin hot aunt you got there

 

[Mr-Fox]

thanks for the link I will look at it

it's just fix one deformed part then another part of the starts to deformed



I'll keep an eye out for one instead of having something like this

 deformed: ass,butt, glans,clothing,arms,legs,feet,breasts,penis,neck,head,wrist,waist,stomach,nose,ears,eyes,hair,fingers,toes,ankles

you just need to put some like <Tag:deformed.01>

it's just a idea don't know if it possible

There are some Lora's and embedding etc for negative prompt.

Embeddings (Textual Inversion).

You must be registered to see the links

You must be registered to see the links

I'm sure I have seen Lora's also but you would need to search for them manually since it's not a conventional category.

 

[Anon1341]

my first work with SD and dreamshaper.
I spent about 4h of learning stuff, generated about 2k images in total :X I hope Ill be better at this....

Does anyone know a good way to inpaint a transparent clothes while keeping the previous things like, e.g. a tattoos and so on?

orginal image promnt:

Spoiler

You don't have permission to view the spoiler content. Log in or register now.

final result:

 

[Elefy]

this is a prompt sharing thread

This thread is for our praise https://f95zone.to/threads/ai-art-show-us-your-ai-skill.138575/

Anon1341 congratulations for your 4h effort and welcome

 

[picobyte]

 

[Deleted member 5469217]

StepSisWasNeverStuck here's another few for you. You'll have to change the model to chilloutmix and rearrange the text.
Probably apply an IG filter on top as well or use phtoshop to change lighting.

Get png info from PNG INFO tab

View attachment 2535885 View attachment 2535886
View attachment 2535900 View attachment 2535894

Oh - also try your luck with the hyperbreasts LORA. And you'll have to compare text with the IG channel to see how you can make it more like that.
Probably enforce the "slim thicc" or prioritise thicc and play with that.

GL. I'm going back to messing with other styles.

I tried to replicate the image on the bottom right but despite copying alll prompts and using the same model, it sill doesnt look quite as yours, how can this be?

 

[Mr-Fox]

I tried to replicate the image on the bottom right but despite copying alll prompts and using the same model, it sill doesnt look quite as yours, how can this be? View attachment 2547873

It could be a setting that you are not aware of and/or a different VAE etc.
It can also be that he is using a different version of the same checkpoint and/or Lora's
This is just the nature of SD, it's not always possible to get an exact replica but you should be able to get something very similar.
A tip is to load the image miaouxtoo posted into PNG info and then "send to text2img". It's a much more convenient way than doing it manually and some settings that are "hidden" in the menu's is going to be added as override settings. Unfortunately not all settings will be transferred. You have to change Face Restoration to Codeformer manually if you have it set to GFPGAN.
Here's my result:

 

[miaouxtoo]

I tried to replicate the image on the bottom right but despite copying alll prompts and using the same model, it sill doesnt look quite as yours, how can this be? View attachment 2547873

Do you have the negative loras? And it looks like you might have changed some of the negative prompts? I can't tell ... I didn't look at mine haha.

 

[Mr-Fox]

Do you have the negative loras? And it looks like you might have changed some of the negative prompts? I can't tell ... I didn't look at mine haha.

I took a look at his image and he didn't replicate the settings, he was not even using hires..
Negative lora? I didn't see any in the data, only the negative tags.
Did you mean negative embedding? "easy negative" I didn't notice this before.

Spoiler: Prompt

You don't have permission to view the spoiler content. Log in or register now.