[Stable Diffusion] Prompt Sharing and Learning Thread

A

atheran

Member
Feb 3, 2020
355
2,753
Sepheyer said:
Adding Lyrics to a Prompt

Just saw someone's promp, he just dropped a few song verses on top of whatever and it cameout nicely. Probably, under the hood, this acts like a seed.

So, I tried adding "Cry Me a River" verse to this red bikini clunker here: link and I gots:


So, yea, when you have an image but you want a tiny difference, as in: same but different, just throw in some random sweet nothings and good things might happen.
Click to expand...
I used a single lyric from a song and got an entire theme out of it. Just changed one word to fit better with what I had in mind. Was probably the first or the second prompt I tried.

00228-1081762041.png

Just the line "all I am is a (((ginger rebellion))), I won't hurt you but I might hurt someone" from the blond rebellion lyrics by Pretty reckless.

Of course it worked with the normal lyrics, I just wanted the subject to be a redhead.

To clarify, my original prompt was just that lyric line. All the rest was added to modify the background and the clothing. Still, both of those things, I got from random renders I did with the prompt, I didn't think to give her a romper initially for example.
 
Last edited:
  • Like
Reactions: Synalon, VanMortis, Mr-Fox and 2 others
Sepheyer

Sepheyer

Well-Known Member
Dec 21, 2020
1,526
3,596
Stealing people's sht:

dfq4l9k-80fb64ac-c862-4063-998c-6e8f430f8274.png

And for science, was curious how this was done, not quite my cup, hehe:
You don't have permission to view the spoiler content. Log in or register now.
 
  • Love
Reactions: Mr-Fox
Nano999

Nano999

Member
Jun 4, 2022
153
68
fr34ky said:
As promised, I tested it and could achieve the results.
View attachment 2351515

To be honest I took all this time to do this because I wanted to understand how to use those Lora, since it's a topic I'm interested in, not to be nice or anything. You only have to follow the instruction the guy give in the model, he says this:

How to use LoRA's in auto1111:
  • Update webui (use git pull like here or redownload it)
  • Copy the file to stable-diffusion-webui/models/lora
  • Select your LoRA like in this video
  • Make sure to change the weight (by default it's :1 which is usually too high)
Go and read that stuff and watch the video he suggests, if you do that exactly you'll be able to use it.

My recommendations:
- download the model he used in combination with the lora, it's called anything v4.5
- Just copy the prompts and other details from this file -> go to png info and drag the picture in your webUI and you'll have everything there.
- update your webUI if necesary, this is very important
- follow what he does in the video
- what he mentions about 'the weights' it's very important too, you can see he edits part of the Lora prompts on the video, that is key.

If you don't touch the weights you will get something like this:
View attachment 2351593
Click to expand...
Links for lora and auto 1111 and anything 4.5?
Where is the video you are talking about?
 
fr34ky

fr34ky

Active Member
Oct 29, 2017
812
2,167
Nano999 said:
Links for lora and auto 1111 and anything 4.5?
Where is the video you are talking about?
Click to expand...
That post is based on a previous conversation in this thread, you should go read all the conversation, or all the thread more conveniently, there is too much context to that comment.
 
  • Like
Reactions: Nano999 and Mr-Fox
Mr-Fox

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,793
Synalon said:
Can somebody PM me the steps on how to upscale an image with Automatic1111 please.
Click to expand...
1. It is very simple. Just press "send to extra" or load picture in the Extra menu.

2. Then either adjust the multiplier how many times you wish to increase the size or choose the resolution to upscale to.

3.Then choose an upscaler, I recommend SwinIR_4x you only need one.

4. Another thing to consider is post processing. If you wish to try it, go to settings and post processing. Chose either CodeFormer or GFPGAN or both, you can adjust them then in text2img and img2img and Extra. Try without , try with either one and try with both and then pick the one you like best.

*I forgot to say that it's best to use post processing when you upscale rather then when you generate images because it introduces yet a variable and you don't know if it's the prompt or the settings or even wich setting etc. Also I believe that it yields better result this way.
 
Last edited:
  • Red Heart
  • Like
Reactions: Sepheyer and Synalon
S

Synalon

Member
Jan 31, 2022
202
626
Mr-Fox said:
1. It is very simple. Just press "send to extra" or load picture in the Extra menu.

2. Then either adjust the multiplier how many times you wish to increase the size or choose the resolution to upscale to.

3.Then choose an upscaler, I recommend SwinIR_4x you only need one.

4. Another thing to consider is post processing. If you wish to try it, go to settings and post processing. Chose either CodeFormer or GFPGAN or both, you can adjust them then in text2img and img2img and Extra. Try without , try with either one and try with both and then pick the one you like best.

*I forgot to say that it's best to use post processing when you upscale rather then when you generate images because it introduces yet a variable and you don't know if it's the prompt or the settings or even wich setting etc. Also I believe that it yields better result this way.
Click to expand...

Mr-Fox said:
I posted in the learning thread also in case it would help someone else.

1 . It is very simple. Just press "send to extra" or load picture in the Extra menu.

2. Then either adjust the multiplier how many times you wish to increase the size or choose the resolution to upscale to.

3. Then choose an upscaler, I recommend SwinIR_4x you only need one.

4. Another thing to consider is post processing. If you wish to try it, go to settings and post processing. Chose either CodeFormer or GFPGAN or both, you can adjust them then in text2img and img2img and Extra. Try without , try with either one and try with both and then pick the one you like best.
Click to expand...
Thank you, with your help I managed to sort this image out finally.

00000.jpg

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

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

You don't have permission to view the spoiler content. Log in or register now.
 
  • Like
  • Love
Reactions: fsen9595, Jimwalrus, Mr-Fox and 2 others
Sepheyer

Sepheyer

Well-Known Member
Dec 21, 2020
1,526
3,596
Synalon said:
What prompts did you use?
Click to expand...
The entire song. The prompt is inside the PNG file - open it with notepad. This is kinda the reason we stopped using "prompt/nrompt" spoilers in this thread because the file contains the prompts. Naturally, then the file has to contain the prompts else it is the thread rules violation.
 
  • Like
Reactions: Jimwalrus, Mr-Fox and Synalon
Sepheyer

Sepheyer

Well-Known Member
Dec 21, 2020
1,526
3,596
  • Like
Reactions: Mr-Fox
Mr-Fox

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,793
Sepheyer said:
Stealing other people's sht. This one using
You must be registered to see the links


View attachment 2445841
Click to expand...
I have not seen "ned" before, looks interesting. He made this checkpoint specifically to be working well with his lora's.
So it's probably worth our time to check out his lora's also.

You must be registered to see the links
 
  • Like
Reactions: Sepheyer
Nano999

Nano999

Member
Jun 4, 2022
153
68
A quick question.
When I go for "blowjob" promts I always get something weird like floating penises without a guy and even when I include some promts where I try to explain AI to not draw floating penises it still draws it xD

Any fix or working promt for an example?

I'm using anything 4.5 model.
 
You must log in or register to reply here.
Top Bottom