[Stable Diffusion] Prompt Sharing and Learning Thread

Sharinel

Active Member
Dec 23, 2018
598
2,511
And there we go! Cheers! Life saver. Will now devote 1000 sleepless hours to playing with one!

View attachment 3501091
If you are running SDXL I would recommend or funnily enough The Pony model -

Pony is set up exactly how it seems (A NSFW checkpoint for my little pony - don't ask) but as often happens the guy that made it was so insane that he basically set up a NSFW base from scratch, which means that it does NSFW stuff a lot better than other models. The only issue is trying to get realistic skins on what is an anime/cartoon checkpoint. I've got it mainly working the way I want though, although best I can do is semirealistic

00248.png
 

felldude

Active Member
Aug 26, 2017
572
1,695
Testing the FP8 training mode that Intel came out with then NVIDIA magically has a version.

Per the paper it is more accurate, I go from 2IT per Second down to 1.33Sec per IT

With no information on optimizer or using Xformers, Debiassed estimation or SNR Gamma, I am relying on testing.
Just based off the loss rate it appears to need more steps then FP16 (Defiantly more steps the brute forcing BF16 with AdaptAdam)
 

nobrumski

Member
Donor
May 27, 2019
172
150
Other than Civitai and huggingface, which I haven't found a good way to browse as easily as Civitai, anyone have any suggestions for checkpoint or Lora download sites or discussion sites specific to models? Not really looking for anything particular, but I know there are models that get removed by users as they delete or close their accounts or Civitai removes them.

I just wasn't sure if there was a good site similar to F95 aimed at SD models and discussion something that has dedicated threads to various models.

Civitai seems to lean way more anime though it is probably just nature of the beast, and looking for more like what we see in the popular 3D style VNs here.
 

felldude

Active Member
Aug 26, 2017
572
1,695
Other than Civitai and huggingface, which I haven't found a good way to browse as easily as Civitai, anyone have any suggestions for checkpoint or Lora download sites or discussion sites specific to models? Not really looking for anything particular, but I know there are models that get removed by users as they delete or close their accounts or Civitai removes them.

I just wasn't sure if there was a good site similar to F95 aimed at SD models and discussion something that has dedicated threads to various models.

Civitai seems to lean way more anime though it is probably just nature of the beast, and looking for more like what we see in the popular 3D style VNs here.
Coincidentally (Actually because I have been working on this for months)
I just posted a XL checkpoint at Civati that has a considerable amount of Daz3d training.

I resisted the urge to do image to image so you can run these prompts and also test them against other checkpoints.

ComfyUI_00075_.png
ComfyUI_00075_.png ComfyUI_00074_.png ComfyUI_00073_.png ComfyUI_00067_.png ComfyUI_00065_.png ComfyUI_00063_.png ComfyUI_00062_.png ComfyUI_00061_.png ComfyUI_00047_.png ComfyUI_00031_.png ComfyUI_00029_.png ComfyUI_00028_.png
 
Last edited:
  • Like
Reactions: sharlotte

BlenderGuy

Well-Known Member
Nov 17, 2023
1,585
3,878
If you are running SDXL I would recommend or funnily enough The Pony model -

Pony is set up exactly how it seems (A NSFW checkpoint for my little pony - don't ask) but as often happens the guy that made it was so insane that he basically set up a NSFW base from scratch, which means that it does NSFW stuff a lot better than other models. The only issue is trying to get realistic skins on what is an anime/cartoon checkpoint. I've got it mainly working the way I want though, although best I can do is semirealistic

View attachment 3502574
Thanks for the heads up! I am also using a site called Tensor these days. Which features upscaling etc.
 

Sgt028

Member
Dec 17, 2020
392
397
Guys, I created a character using AI. Is it possible for me to use this character in several sex scenes? Every time I tried, the AI changes my character, and I would like to use it without wanting to change the character's characteristics. image.jpg
 

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,802
Guys, I created a character using AI. Is it possible for me to use this character in several sex scenes? Every time I tried, the AI changes my character, and I would like to use it without wanting to change the character's characteristics. View attachment 3543805
Controlnet with openpose for the pose. You can try using ip adapter to try retain the likeness of the character if necessary.
 

XenoXe

New Member
Sep 8, 2021
10
4
Guys, help me out on this.
Is there any good tutorial on fixing the hands on SD1.5 for both the manual way using photoshop and using the inpainting tools for from the sd webui.[For Anime images only]
 
  • Like
Reactions: namhoang909

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,802
Guys, help me out on this.
Is there any good tutorial on fixing the hands on SD1.5 for both the manual way using photoshop and using the inpainting tools for from the sd webui.[For Anime images only]
The genre doesn't matter, hands and toes are a constant issue with SD among other things.
There have been attempts to fix it with varying results. There isn't any perfect solution that works 100 percent of the time unfortunately. However, there are of course ways to improve the result and get something that doesn't look like a mutant.
Best practice is to try avoid bad hands to begin with.

Use a good resolution ratio.
Since you are using SD1.5 this means you need to keep the resolution under 1024 for the sample image otherwise you will likely get 2 merged images, resulting in mutants and aberrations. I would suggest using 512x768 (2:3 ratio) alternatively 576x768 (3:4 ratio), 640x960 (2:3 ratio) alternatively 720x960 (3:4 ratio) if you want a bigger image.
In my experience you get better results with a portrait ratio (2:3 or 3:4) rather than letterbox ratio (square 1:1), this might depend on how the checkpoint has been trained though.

Don't use too many loras or embedding at once.
When a lora or embedding is being trained, the entire subject is always being trained. This means that there is no such thing as only an outfit lora for example even though people claim this. So when you use many loras and/or embeddings you are more likely to have clashes that will result in mutants.

if you are using openpose, take the time to refine the pose in the editor especially the hands and fingers. The preprocessors are not even close to being perfect in it's detection.
Don't keep banging your head against the wall and trying to make a checkpoint model do something it has not been trained for or simply can't. If you keep getting bad results try several different checkpoints and see which gives the best result. For this purpose xyz plot script is great.

Go through the prompt and simplify it, remove any clashing words.

Most of the time I avoid using inpainting and simply generate more images until I get something good, but sometimes it can't be helped. After detailer has a hands model the will do inpainting for you on the fly, this doesn't work flawlessly but it can improve the result significantly when it works. I highly recommend getting adetailer and learning how to use it.

I have made tutorials exploring hands issues with inpainting manuallly. It's not perfect but it will at least point you in the right direction.
Other option is using the inpainting type plugin for the free image editor krita. I will link to this as well.

XYZ Plot Script Tutorial: https://f95zone.to/threads/stable-diffusion-prompt-sharing-and-learning-thread.146036/post-12816116

Fixing Hands with inpainting: https://f95zone.to/threads/stable-diffusion-prompt-sharing-and-learning-thread.146036/post-12545892

Krita Tutorial: https://f95zone.to/threads/stable-diffusion-prompt-sharing-and-learning-thread.146036/post-12643242
 
Last edited:

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,802
Guys, help me out on this.
Is there any good tutorial on fixing the hands on SD1.5 for both the manual way using photoshop and using the inpainting tools for from the sd webui.[For Anime images only]
Ps*

If you want better and more specific help you need to post an example image with the generation data included so we can take a look at the prompt and settings etc. We can then help you better with trouble shooting. Also give more descriptive info about what your goals are so we know what to aim for.
 
Last edited:
  • Heart
  • Like
Reactions: DD3DD and Sepheyer

felldude

Active Member
Aug 26, 2017
572
1,695
I'm at 300+ trainings and today I had one of my worst trainings followed by one of my best.
The worst I knew was going to be bad, the image quality sucked, angles where not there.

But now for the best, I'm still amazed it is producing the results it is.

The 512x512 proof of text to image
ComfyUI_02007_.png

1024x1024 with a Lora value of 1.0 and Clip strength doubled at a full image de-noise of 0.7
ComfyUI_02008_.png

1536x1536 full image (Not Highrez fix) de-noise @0.5 on SD 1.5 not XL
ComfyUI_02014_.png
 

felldude

Active Member
Aug 26, 2017
572
1,695
Considering it was trained on 112 Images and fairly low rank....I am still in disbelief that it consistently able to go up to 1536x1536

ComfyUI_02018_.png ComfyUI_01947_.png ComfyUI_01945_.png

ComfyUI_02019_.png
 

osanaiko

Engaged Member
Modder
Jul 4, 2017
2,476
4,474
I recently tried SDXL after lots of messing around with SD1.5

However I find that the controlnet openpose does not seem to work very well.

I did a "git pull" update in the "stable-diffusion-webui\extensions\sd-webui-controlnet" folder to get sdxl support.

And downloaded a model "controlnetxlCNXL_bdsqlszOpenpose.safetensors" which seems to be the right one?

Do i need to get the " " model as well (or instead?) (it is very large)