Software Engineering Courses (SECourses)•3y ago

use sdxl you will get much better images

FFurkan Gözükara SECourses use sdxl you will get much better images

vKorby•10/23/23, 11:30 PM

Oh ok thank you, is this it?

FFurkan Gözükara SECourses it is a so lame model

vKorby•10/23/23, 11:31 PM

yeah its jus he used it and it goes viral back to back

VvKorby Oh ok thank you, is this it?

Furkan Gözükara SECoursesOP•10/23/23, 11:40 PM

ye it is a good one

VvKorby Hey guys jus joined in here, I've been trying to figure out if this is stable di...

EyeSpyBekoAI•10/23/23, 11:53 PM

Chiming in. Yes. I’m aware of that channel I was watching a few of their vids too. Definitely a SDXL based model. I forget which video but there were one or two that had pretty distinct style that I also see a lot in my generations. If you’ll notice also the guy in his videos prompts a pic based on his narration and puts them together as slides as videos. It’s a nice way to get a short without doing a full vid.

EEyeSpyBekoAI Chiming in. Yes. I’m aware of that channel I was watching a few of their vids to...

vKorby•10/23/23, 11:58 PM

Yeah, heres another vid

vKorby•10/23/23, 11:59 PM

he's the only one doing these type of images, running multiple accounts. Ik him but he gatekeeps lol

EyeSpyBekoAI•10/24/23, 12:00 AM

Yup. Sdxl based model. They have a bit of an anime style if it’s not sdxl base directly I’m sure you’ll find it on civitai. There aren’t too many sdxl based models yet.

VvKorby he's the only one doing these type of images, running multiple accounts. Ik him ...

EyeSpyBekoAI•10/24/23, 12:01 AM

I didn’t realize he was going viral. I thought I was seeing them because I watch AI based content constantly so figured that’s why I was seeing them

EEyeSpyBekoAI I didn’t realize he was going viral. I thought I was seeing them because I watch...

vKorby•10/24/23, 12:03 AM

Yeah lmfao, we have a group for brands that we promote. He's the only one doing these type of images, took me a whole month too figure out he's using SD. Bouta try it rn gonna see if i can get sum similar

Furkan Gözükara SECoursesOP•10/24/23, 12:08 AM

i dont think sdxl

Furkan Gözükara SECoursesOP•10/24/23, 12:08 AM

images are so low quality

FFurkan Gözükara SECourses images are so low quality

vKorby•10/24/23, 12:09 AM

what do u think it is?

VvKorby what do u think it is?

Furkan Gözükara SECoursesOP•10/24/23, 12:15 AM

i dont know which sd model

Furkan Gözükara SECoursesOP•10/24/23, 12:15 AM

but it is sd for sure

Furkan Gözükara SECoursesOP•10/24/23, 12:19 AM

i am going to sleep hopefully see you tomorrow

Furkan Gözükara SECoursesOP•10/24/23, 12:19 AM

tenssort public tutorial done

Furkan Gözükara SECoursesOP•10/24/23, 12:19 AM

uploading

VvKorby what do u think it is?

EyeSpyBekoAI•10/24/23, 12:26 AM

Hey, my bad if I mislead you. @Dr. Furkan Gözükara is right that’s not SDXL. Some of the writings that came up is a giveaway. I watched the videos you sent in full. In fact, I don’t think this is the same person doing the videos I mentioned, this guy is hawking snake oil products. He might be copying the other vids I mentioned. The ones I saw talk about history facts and stuff. Unless it’s the same guy having multiple channels.

EEyeSpyBekoAI Hey, my bad if I mislead you. @Dr. Furkan Gözükara is right that’s not SDXL. Som...

vKorby•10/24/23, 12:28 AM

yeah its the same person running multiple different products

vKorby•10/24/23, 12:29 AM

but if u saw some vids its prob from dif people, like using midjourney or leonardo.ai

vKorby•10/24/23, 12:30 AM

his vids are different, cuz its like more vibrant and detailed

vKorby•10/24/23, 12:31 AM

hard to explain but you can tell

VvKorby hard to explain but you can tell

EyeSpyBekoAI•10/24/23, 12:33 AM

I found the channel I was talking about. Yea the guy you posted is stealing the style of the person I mentioned. Will DM you the video so we don’t clog up this channel.

EEyeSpyBekoAI I found the channel I was talking about. Yea the guy you posted is stealing the ...

vKorby•10/24/23, 12:34 AM

bet bet

t800se•10/24/23, 11:27 AM

Has anyone tried to use Realistic Vision v5.1 with their own trained models? Am I missing something? I trained with my own images and created some Loras. However, I can't use them when I pick Realistic Vision in the checkpoint selector. Do I need to use some other model when training (to be able to use Realistic Vision)?

Tt800se Has anyone tried to use Realistic Vision v5.1 with their own trained models? Am ...

Furkan Gözükara SECoursesOP•10/24/23, 11:52 AM

no it should work

Furkan Gözükara SECoursesOP•10/24/23, 11:53 AM

loras wouldnt work only when sdxl vs sd 1.5 based model

Zono50•10/24/23, 12:25 PM

What's the recommended way to train a 1.5 lora on your face now?

Zono50•10/24/23, 12:32 PM

@Dr. Furkan Gözükara

iamannakogan•10/24/23, 12:56 PM

Where can I find recommendations on how to select photos for SDXL pipeline: https://www.patreon.com/posts/89213064 ? I used 14 photos but the results don't look like me - I suspect my photos might be off.

mikemenders•10/24/23, 1:24 PM

Your photos will be 1024x1024 resolutions, use minimum 20-25 with different places, lighting etc.

Iiamannakogan Where can I find recommendations on how to select photos for SDXL pipeline: htt...

Benjamin•10/24/23, 1:29 PM

see also as a reference his training data set for the post you linked... with only 13 images

Khang•10/24/23, 2:29 PM

@Dr. Furkan Gözükara Hi Dr.Furkan & everyone, is there any guide for training ControlNet models (for non-coders)? I found a ControlLoraSave node but not sure how to use it. Is ControlNet training much harder than Lora training, so only a few folks/companies can do it?
https://github.com/Stability-AI/stability-ComfyUI-nodes & https://github.com/HighCWu/ControlLoRA

akellio•10/24/23, 3:04 PM

Hi, who has tried running on nvidia A100? What is the fastest video card for sdxl lora training on runpod?

iamannakogan•10/24/23, 5:01 PM

Thank you!

Mmikemenders Your photos will be 1024x1024 resolutions, use minimum 20-25 with different plac...

iamannakogan•10/24/23, 5:02 PM

Thanks!

rafraf•10/24/23, 5:55 PM

@Dr. Furkan Gözükara is there a tool to find duplicates within a large dataset?

mikemenders•10/24/23, 5:56 PM

Speaking of Adafactor: why can't I select the Adafactor scheduler if the optimizer is Adafactor? I get an error and it suggests to use constant with warmup optimizer.

Mmikemenders Here is my config for style. I used 10 images with 10 repeats, captions and 10 e...

Jas•10/24/23, 5:58 PM

Hey, I see you are very knowledgable about creating 1.5 LORAs.

I am trying to create some LORAs based on art styles/artworks.

Would this .json be appropriate?

I have got 50 images I want to train with for each style, or do you recommend less?

Is the base 1.5 model best to use? I have seen others recommending RealVision.

I am testing with training images 768x768, although I see you recommended 512x512, will that be ok?

Cheers for your help

mikemenders•10/24/23, 6:12 PM

@Jas The large file is also good for styling, and can be reduced to 512 by lora.
Which is why I train on SD 1.5:
- 512 resolution will take over the style, Hires can scale up the result, plus it adds exciting new detail, so it's okay if it burns in a bit.
- If you don't have powerful hardware, 512 will do. 768 will take longer to train, but you can also train on SDXL.
- I got one of the best training in style on Prodigy. Plus with 10 images, so 50 images will be perfect, even with 10 epochs.
- Label your pictures, but not with danbooru, but with a one sentence description of the picture in natural text, but not too long. If you use WD14 tagging, you will still have style, but SD 1.5 interprets it a bit differently and doesn't know everything from the words, so you may slip a bit on style. It won't be bad, but it will be a bit different. SDXL is supposed to understand the text better now.
- Don't use reg images, I recommend 10x repetition.

mikemenders•10/24/23, 6:14 PM

Here is my json for styles

Mykee_-_Prodigy_-_SD15_-_style.json2.97KB

mikemenders•10/24/23, 6:15 PM

One more thing in favour of SD 1.5: if you train your style on it, you can mix the lora with most models, and you can get very good results with cartoon models.

vKorby•10/24/23, 6:16 PM

Hey guys, i need some help here does anyone have any videos or advice if my SD is running literally for like 30 mins lmfao

mikemenders•10/24/23, 6:22 PM

Sometimes Auto freezes for me too, so restart the backend cold (close command prompt and restart it).
Use the --medvram-sdxl option. The preview image will be poor, but you will get a good image at the very end of the generation.

Mmikemenders Sometimes Auto freezes for me too, so restart the backend cold (close command pr...

vKorby•10/24/23, 6:37 PM

ight bet, will i still get the same type of images? because im looking for an anime/realstic type of image

VvKorby Yeah, heres another vid

vKorby•10/24/23, 6:37 PM

like this

mikemenders•10/24/23, 6:39 PM

Of course. The --medvram-sdxl option is there to generate the same image faster and in a different way. And it really will generate faster, and you'll see the end result at the very end.

Mmikemenders Of course. The --medvram-sdxl option is there to generate the same image faster ...

vKorby•10/24/23, 6:42 PM

sorry kinda new to this, where is that option under?

mikemenders•10/24/23, 6:55 PM

Type it to webui-user.bat file after set COMMANDLINE_ARGS=set COMMANDLINE_ARGS=

mikemenders•10/24/23, 6:56 PM

like this: set COMMANDLINE_ARGS=--medvram-sdxlset COMMANDLINE_ARGS=--medvram-sdxl

use sdxl you will get much better images

Similar Threads

Similar Threads

Similar Threads