Видео

Is the cat real?

I haven't found anyone that has a fix for Safari browser.

I've tried and tried to get this program to work but it wont

in negative prompt ofcouse

((6 fingers++)), this made it correct i think

too many hands seems ? i dont know if it has any significance in result , hast tryed out, didnt have prblm with model with to many hands or lims, exept 1 picture , she sits in Lotus posistion , and thers 3 feets, had to erease and clone tool in photoshop in like 6/10 images becouse the model are hard to be told that a human has 5 fingers. But still Krita ai and stable diffusion Comfy/ui is the greatest invention since the invention of fire !? Sleep well and have woanderfull dreams ;) Tomorrow is an important day allways ;)

negative prompt " copyrighted : drawing, painting, morph, artifacts, sketch, bad quality, low resolution, blue dress, blurry, BUTTERFLYES,, deform limbs, child, teenager, ((6 fingers++)), (3 legs) ( 1 leg), to many fingers, man, to few fingers, cartoon, disney, low quality, bad lightning, anime, artifacts, to many hands, retro, gibli,

had 6 fingers corrected in 8/10 try, after providing negative prompt after your direction, phifani no21, Thax Streamtabulos ;)

ruclips.net/video/U9OJoYa2JRA/видео.html, Zaxious_XL Lightning Model can make great videos in Haiper ai as Haiper ai 1.5 and stable diffusion unerstand eachother ? Have consiatent problems about the ZaxiousXL_lightiningV30 model as it 9/10 made 6 fingers and some times 3 legs and so. Your braggets and +++ system made me have a lot better control over floating items and artifacts. Thank you again for sharing your knoledge for all us neewbies ;) Its absolute top noch education in a time where all is fake :) Tx ;)

i complained to youtube and i don't have it anymore!!

krita ai is all i need.

Hi, Any solution?? I get this error: Error during server startup: OSError: [WinError 126] The specified module cannot be found. Error loading "C: \Users ame\AppData\Roaming\krita\ai_diffusion\server\python\lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies. [1]

9:42 Thank you for like 20 epiphany’s. So many ossiers i had with Photoshop vs new choise , use Krita Ai now and the Zarcasious Lightning model. Amazing resultat with a red haerd Godess 🥳🥳. Inspiration Lakshmi India Godess Alssund PictureThis with Lotus flowers 🥳👍🤓

Hello. When I add the downloaded styles to the checkpoints folder and hit refresh, the styles don't appear in the dropdown. I've tried restarting Krita to no avail. I'm using AI plugin version 1.19.0 and Krita 5.2.2 Thanks in advance. Cheers!

You don’t need that many images for a single concept/subject finetuning. You could easily keep it at 1/3 of that and good results. With that said: thanks for sharing onetrainer settings. I work with support from RunDiffusion on Runpod since a year but it’s nice to be able to try this out on my 2070 super soon.

I'm late to the party, but I didn't have the annoying new layout since only recently. Thank you for this tutorial, the new layout is ridiculous and I wonder if RUclips want's us all to stop watching. Very happy to have the old style back.

it says i have no tokens??

Есть полезная информация, однозначно есть, но автор очень много любит чесать языком. Контента на 5 минут, но языком почесал рассусолил на 40 минут

where did you upload your image?

thank you so much for the SDXL json. Please do share SD1.5 Jason too and a video if possible :) awaiting for your response

i dont have SDXL - lighting Dreamshaper.. where to get it plz ?

Why does mine takes a long time to generate it's images, i have a RTX 4070 and it's taking forever, even after increasing it's power in the settings :c

Is it mandatory to install all modules or I can instal modules which I prefer?

I have trained with this setting on my rtx-3060 8g vram. Its training speed is 1xx s/it (batch 4). Only takes 2 s/it (batch 1) on Kohya_ss. If I need to train LoRA to 5000 steps, it will take over 50 hours to train. I don’t think this speed can “alternate” Kohya_ss

Ok, I followed all the hints from the three videos, I also installed all the software as suggested by @SECourses (Python, Git, CUDA, etc.), but I still get the "CUDNN_STATUS_NOT_SUPPORTED (....Conv_v8.cpp:919.)" error. Any idea how to fix this? Should I change CUDA version (I installed CUDA 11.8)? Should I change Pytorch version (I have 2.3.1)? Python is v.3.10.11. I run OneTrainer on Windows 11, with a RTX 4070. Thanks in advance to anyone willing to help me.

Hi, Your channel is being quite valuable to understand Krita and AI. Thanks a ton! Do you know if it's possible to select (keep) the dress and redo the inverted selection?

Great videos about Onetrainer, I have watched the first two (will watch the third later on). I am learning a lot from you, thanks!!! I have the CUDA error too (UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR: cudnnFinalize Descriptor Failed cudnn_status: CUDNN_STATUS_NOT_SUPPORTED (Triggered internally at ..\aten\src\ATen ative\cudnn\Conv_v8.cpp:919.) I am running Python 3.10, and tried to install latest CUDA (12.5.1) on Windows 11 64bit. I checked around, and maybe it's something about what version of Pytorch I am running... The Pytorch 2.3.0 had some problem with latest CUDA, but 2.3.1 should have fixed them and should be fine with CUDA 12.1. Problem is I have no idea how to change Pytorch (or even check what version I am running) and how to uninstall CUDA 12.5 and install the older 12.1... Anyway, trying to create my first Lora right now! Thanks again!!!

Strange, I have a 1070 gtx and models are running from a ssd. Copied all your settings and my generation times are much higher (02:30 for one image)

I'm just getting started on my Krita journey, thanks for making it relatively easy!

16 minutes to tell me two minutes of what to do. I simply got jack of listening to someone run their mouth and went elsewhere.

You cannot see the comments in theatre mode, absolutely stupid. I log out of google and the original format returns till I log into google. I'm off to TikTok.

There is it for Brave??

640x640 or 704x704 may work better than 680x680 as stable diffusion likes to work in multiples of 64, you might see less artefacts

I just want to know the absolute reality you have. how to have like that one? thanks

wait....can I import a videoclip finished I ve made before, and on krita, select for example some frames and select and change my face by a vampire face (for exemple) with consistence during some frames?

Screw youtube any company with half a brain would of added an option so people could choose between the old layout or new layout, i rarely use youtube anymore since they changed it to the new annoying layout, all they are doing is pushing people off their platform to go elsewhere, end of the day old web search gives more variety too..

I'm showing my ignorance here. But after weeks of using Instant ID with absolutely superb results. Add a face, type in a prompt. 15 seconds later an image that is almost perfect of that face that works at almost work all head angles in different lighting etc. Then looking at what these model trainers require to achieve the same results. A degree in rocket science, a £1200 graphics card, hours of training on a large image set - and... after doing it, results that as SO much worse than Instant ID I'm wondering why the best results I get are from the simplest, cheapest and quickest method. (by a country mile) What is it that Instant ID does that all this faffing about with Loras fails to even get close to?

and if you trace it, it can be really be yours!

Way to complicated to install on a Mac, not worth the time

Krita/ComfUI cannot handle if you ask it for a plain background or to remove the background after you've selected the subject. I don't see how it's ever going to compete with Adobe.

U are so great, love ur work

I hate this soo much...

Hello. Which category would I click in the feedback section? It doesn't show a category to cover this subject.

obrigado pelo video!

Thanks, useful to me

Hi! Thank you so much! the chrome extension worked perfectly for me!!

Amazing! thank you so much! 😄

I am trying to figure this stuff out. I want to use this to put me in dresses I haven't actually sewn yet. Or maybe some clothing inspiration? Is that possible?

First guide on how to properly setup the folders for kohya to run, love it!

Thank you thank you thank you!