[Stable Diffusion] Prompt Sharing and Learning Thread

[Markbestmark]

I merged the Schnell and Dev FP8 models and made these renders as an experiment.

Spoiler: Merge

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

Edit: Randomly Added new images.

do you have an examples of animations for those pictures? Have you tried to make some? I am searching as much as I can to make vid2vid or img2vid that would suit to start making games =D So would be great if you can share some tips =P

 

[Synalon]

do you have an examples of animations for those pictures? Have you tried to make some? I am searching as much as I can to make vid2vid or img2vid that would suit to start making games =D So would be great if you can share some tips =P

I haven't tried to animate yet, I'm not good at creating workflows and Flux uses some nodes I haven't seen before so I don't know how to integrate them into an animation workflow yet.

If somebody is willing to make a workflow with flux that can animate and share it I will give it a try.

The alternative I have is to just load one of those images into animated diff and use another checkpoint to animate, and that will probably lose the clarity.

 

[Markbestmark]

I haven't tried to animate yet, I'm not good at creating workflows and Flux uses some nodes I haven't seen before so I don't know how to integrate them into an animation workflow yet.

If somebody is willing to make a workflow with flux that can animate and share it I will give it a try.

The alternative I have is to just load one of those images into animated diff and use another checkpoint to animate, and that will probably lose the clarity.

that basically what I was doing with my animation =D so that's why my results are not so clear as your pictures. I hoped that someone here would actually help me to achieve better quality =)

 

[Synalon]

This is what I've been using.
Workflow included if you want to try and add animate diff to it.

 

[Onetrueking]

Is it possible to train lora style on these images if i have a lot of similar? Don’t know much about training lora, but i’m out of ideas what lora i need to mix with concept art twilight to get something like that.

 

[felldude]

You must be registered to see the links

This LORA could use testing and an X,Y,Z Plot

 

[felldude]

You must be registered to see the links

 

[ManlyManperorOfMankind]

Liara test

 

[felldude]

You must be registered to see the links

 

[felldude]

With the NF4 model I am able to generate images at 7-12 Seconds per it
This is 10x faster then I was getting out of the FP8 model

If you have a lowvram card and use Comfy UI I highly recommend this for FLUX
Just google NF4 for COMFYUI

EDIT: NF4 has issues with LORA's but it should allow for older card users to use FLUX at a decent speed when QLORA conversion is available lora's should also be usable.

My Shenell NF4 Seconds per IT - 4.64 per IT
My Dev NF4 Sec Per IT - 30 or so

Still a vast improvement from the 100 seconds per IT I was getting.

I tried adding

quantization_config = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_compute_dtype=torch.bfloat16)

along with

import torch
from transformers import BitsAndBytesConfig

But I am still unable to load BF16 or FP16 lora's along side a NF4 model, hopefully a programmer can get a working script for NF4

 

[theyakuzi]

https://f95zone.to/threads/dungeon-...reedom-elburn-v1-12-drill-drill-drill.186493/

how do i achieve something like this, i assume this wasn't created in stable diffusion?

 

[Meyer1234]

Can someone Tell me how do i describe two Charakters in one prompt? Wanna have a picture of a Male and female and describe them. But i Always get a Male with breasts oder female with dicks. Is there a strict Syntax how to apply a specific description to only one of the Characters and another description for the other Character?

 

[osanaiko]

Can someone Tell me how do i describe two Charakters in one prompt? Wanna have a picture of a Male and female and describe them. But i Always get a Male with breasts oder female with dicks. Is there a strict Syntax how to apply a specific description to only one of the Characters and another description for the other Character?

Regional Prompting extension would be a good start.

 

[Sharinel]

Can someone Tell me how do i describe two Charakters in one prompt? Wanna have a picture of a Male and female and describe them. But i Always get a Male with breasts oder female with dicks. Is there a strict Syntax how to apply a specific description to only one of the Characters and another description for the other Character?

as osanaiko said Regional Prompter is good. You could also try using BREAK as that helps. So if using Forge/A1111 try "scene description BREAK female character description BREAK male character description.

I find that SwarmUI does that a lot better than Forge, but there it is <break> as opposed to BREAK.

 

[CBTWizard]

https://f95zone.to/threads/dungeon-...reedom-elburn-v1-12-drill-drill-drill.186493/

how do i achieve something like this, i assume this wasn't created in stable diffusion?

They probably used an SD1.5 model for the characters and either a LoRA or an SD extension like Pixel Art to pixelate it.

EDIT:I made something similar using MeinaAlter v3 as the 1.5 model, with the Thicker Lines LoRA and Pixel Art extension.
I scaled it 2x with R-ESRGAN 4x+ Anime6B and a denoise strength of 0.42069 (yes) with Pixel art downscaling it 4x and a color palette size of 128

Spoiler

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

Regional Prompting extension would be a good start.

as osanaiko said Regional Prompter is good.

For (re)Forge, you can use Forge Couple as a simpler alternative.

 

[zapzapzapzap]

I only know the bare basics of using local stable diffusion and i want to ask, are we in the technological level to uncensor game assets such as Live2D? Not talking about video files, only baked 2d assets. I think it's worth a guide if that's the case.

 

[Sharinel]

I only know the bare basics of using local stable diffusion and i want to ask, are we in the technological level to uncensor game assets such as Live2D? Not talking about video files, only baked 2d assets. I think it's worth a guide if that's the case.

Do you have an example asset? I don't know squat about live2d

 

[zapzapzapzap]

Do you have an example asset? I don't know squat about live2d

Random Live2D asset from a doujin game. What i mean is using inpainting to uncensor, maybe. Currently, uncensoring assets like this is a desired task for a lot of people here

 

[Sharinel]

Random Live2D asset from a doujin game. What i mean is using inpainting to uncensor, maybe. Currently, uncensoring assets like this is a desired task for a lot of people here

You could uncensor it really quickly, the problem is getting the styles matched up with the rest of the image. Literally took 30 secs, just masked out the 2 purple dicks, prompted for 'purple dildo' then did the same with the penis

 

[zapzapzapzap]

You could uncensor it really quickly, the problem is getting the styles matched up with the rest of the image. Literally took 30 secs, just masked out the 2 purple dicks, prompted for 'purple dildo' then did the same with the penis

View attachment 3998260

that's actually impressing, what model?