- Mar 11, 2023
- 46
- 132
Send the instagram link or forum that you got those images from.
They look pretty damn professional - I suspect photoshop might have been involved or a different type of model.
They look pretty damn professional - I suspect photoshop might have been involved or a different type of model.
- Apr 4, 2023
- 12
- 15
Here is the insta ID:-
You must be registered to see the links
yes, he has mentioned photoshop in his bio, how is he using photoshop in creating these images?
- Mar 11, 2023
- 46
- 132
I'm guessing he's using it to enhance or change certain parts. Filters possibly after generation?
It depends if he's an artist or not.
- Apr 4, 2023
- 12
- 15
In that case can you give some tools that will make me achieve the those results.
thank you for telling me how to write the prompt but in order to get results as close as possible I think I need more tools.
can you suggest and provide links for those tools.
- Mar 11, 2023
- 46
- 132
If I were you, I'd learn how to use photoshop. And stable diffusion from the looks of things.
What have you managed to do so far?
What have you managed to do so far?
- Apr 4, 2023
- 12
- 15
Created this with the help of your prompt but it doesn't look that good.
i have not used any upscaler because low vram. something looks off but i can't exactly point what.Can you help?
Last edited:
- Jan 24, 2020
- 1,401
- 3,802
Instead of only saying low vram, it's much more useful for anyone trying to help to specify it. So we know exactly what you are working with and can give proper guidance.
Don't be vague about anything in a learning thread.. Especially not when someone is as generous with his time as miaouxtoo has been. Other's will also read this and learn from it.
Also you never know, you might be able to get much more out of that potato than you think.
Last edited:
- Sep 15, 2021
- 1,045
- 3,994
Even 'upscaling' by one pixel in each direction can give useful changes as it's such a different process to the initial Stable Diffusion iteration.
And, as Mr-Fox says, there's 'low vram' and LOW vram.
4GB will actually allow you some upscaling.
2 or 3GB won't allow much beyond initial generation.
6GB is becoming ever more useable with improved extensions. We're not at the 'Dreambooth-on-6GB' stage yet, but it's certainly been made to work on 8GB.
- Apr 4, 2023
- 12
- 15
I Have Nvidia Quadro T1000 8GB GDDR6 but for some reason i was getting artifacts using Hires upscaler so i didn't upscaled it. But i didn't wanted to talk about the artifacts because the process of getting the image using his prompt was more important then the upscaled image so i used the term "low vram" to not bring this thing up.
I am also here to learn about the process of creating these artworks and i am not here to waste anyone's time.
Here is the upscaled version
- Jan 24, 2020
- 1,401
- 3,802
I never thought you're a time waster. My only point is that if people who wish to help has more information the better aid and guidance they can give. Making good images with SD is a bit like baking a cake. There are many ingredients that makes the cake and each ingredient plays it's part. You get the best result with using the best possible ingredients but also the best ratio between them. The first thing is the prompt, everything starts here, then it's the sample image settings, then hires fix and post processing and restore face etc. After you get the best possible "cake" you can go to photoshop and correct anything that is not as you wish and/or add some "decoration" that will make this cake into a masterpiece. Me and many other's have posted tips and guides about how to make a good image, I highly recommend that you read through them. You can find links to some of these posts on the first page. The same recipe doesn't work for all cakes obviously, so it will change depending on what you are baking but there are similarities. All cakes needs eggs, sugar and flower... It's the same with generating images with SD. All images needs a good prompt and good sample settings, and hires fix take it to the next level. Imo hires fix is not optional, it's required.
Last edited:
- Mar 17, 2021
- 323
- 1,093
About low ram just read this post. I'm on 6 Gb and it work via upscaler and for generating big image, also my
You must be registered to see the links
show it without hi.res fix. This is really helpful for low vram.
- Apr 4, 2023
- 12
- 15
sorry if i sounded rude but i am very new to this and whenever i try to create something in SD it just becomes wierd so i thought you guy can give me a perfect formula to create the image that i want.
But now i know that without me experimenting i am not going to get any where.
So if you guys help me reach my goal(the type of images want to create) i would really appreciate that.
- Apr 4, 2023
- 12
- 15
- Jan 24, 2020
- 1,401
- 3,802
No, I didn't perceive you as rude, I was only clarifying myself. Don't worry about it.
This is the learning thread, we are all glad to help. Yes unfortunately there isn't a quick trick or fix to this. But the fast track for anyone new is available, by reading the guides and tips from people who knows what they are talking about. On the first page of this thread seph who created this thread have linked to some of the posts that has guides and tips, though there are more than is linked to there. So look through the pages and you will find a ton of good information and also links to good resources etc.
- Mar 24, 2019
- 1,102
- 3,982
Played a bit with LoRas imitating Tony Taka / sciamano240 and happoubi jin's styles. Anatomy needs some work....especially fingers/hands. Used together with Anythingv5.
Last edited:
- Mar 11, 2023
- 46
- 132
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
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.
Probably apply an IG filter on top as well or use phtoshop to change lighting.
Get png info from PNG INFO tab
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.
- Apr 4, 2023
- 12
- 15
You are a GOD at this man. I was also playing with the prompt to get the job done but now i think i have to do some experiments with the checkpoints too. Thank you bro
And one more thing the artwork you created has very realistic body but in his artwork it is more like exaggerated body curves, is their anyway to add that look
I feel like ReV Animated is the checkpoint which comes the most to his checkpoint.
link:-
You must be registered to see the links
Last edited:
- Jan 24, 2020
- 1,401
- 3,802
Others to consider:
You must be registered to see the links
You must be registered to see the links
You must be registered to see the links
Last edited:
- Mar 11, 2023
- 46
- 132
I've tried Kotos and Perfect world.
I suspect he might be running it through two? Or have a mix.
--
Ah - he's using img2img after generation.
In his reel that he posted 3 days ago - it shows him adding parts / fleshing out the initial generated image, then he runs it through img2img after, and changes it. He adds the wrist guards it seems?
Is there a particular reason you want to copy this style?
--
For everyone else too, this is a great reference:
Can someone pin that?
I suspect he might be running it through two? Or have a mix.
--
Ah - he's using img2img after generation.
In his reel that he posted 3 days ago - it shows him adding parts / fleshing out the initial generated image, then he runs it through img2img after, and changes it. He adds the wrist guards it seems?
Is there a particular reason you want to copy this style?
--
For everyone else too, this is a great reference:
You must be registered to see the links
Can someone pin that?
- Mar 11, 2023
- 46
- 132
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:
Start from bikini. Throw into PS. Scribble in straps, wrist stuff, whatever. Img2img. Find one you like.
Throw into PS. Scribble. Img2img.
Find background
Pull person from old img2img.
Add her into background
Find flames from somewhere. add in.
Img2img to blend.
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
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:
Start from bikini. Throw into PS. Scribble in straps, wrist stuff, whatever. Img2img. Find one you like.
Throw into PS. Scribble. Img2img.
Find background
Pull person from old img2img.
Add her into background
Find flames from somewhere. add in.
Img2img to blend.
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