Видео

add only this command --disable-cuda-malloc work for me. work in 1024x1024 resolution with fp32 very fast. my gpu 1660 super 6gb

Has anything changed about this. I am trying this on a 3070Ti (8Gigz) with your config after changing to bf16 and seems like its taking about 90 minutes per epoch for a 100 image Lora. Most images lower than 800p res, and a few under 1500p.

I have successfully trained a Lora on my own face and the samples looked really good, but for some reason no matter what I do on stable diffusion, it generates an image of some other person and not me, not even close to me actually. I don't know what I'm doing wrong, I've tried using all sorts of trigger word found in the captions and even used the name of the concept but it doesn't work. Can anyone help me?

Hey, I just wanna tell you that finding your channel is like a diamond to me. It may change my life. A lot has been going on in life. 19 yo didnt graduate high school due to health reasons and bla bla bla. And im going to take a shot in this kind of field. Whatever happens to you, I just want to express my grateful for the amazing knowledge you share.

Excellent, beaucoup de choses très intéressantes, merci beaucoup et bonne journée, je m'abonne... Excellent, lots of very interesting things, thank you very much and have a nice day, I'm subscribing...

1:40 Speed running "video is 40 minutes"... just how difficult this is going to be... Pain... 30:41 the skip here to a "finished product" was terrible, no clue how to continue after getting the first samples (that are inaccurate) and how to formalize the working session into a file, no info into how to save the training data or anything really, I know is a free tutorial but it feels more like self promotion.

Thanks!

Hi. I want to use the same 5 prompts on 10 different pictures. How can I do that without having to do the same stets again on each pictures?

how to turn on the negative prompt in krita? thank you Bro

If no one has said this yet, take off your glasses when taking pictures of yourself for AI. Same if you wear hats, scarves, anything that covers the part of your body that you want the model to learn. Glasses can be added to the image during generation. Alternatively, you could have a set of images specifically in which you wear the glasses, and adjust your head in position and angle relative to the camera, and in different light settings, then add those to the dataset while specifically tagging things like lighting, position, and angle. As an alt alternative, you could use those images to create a second (glasses) lora to use with the first, and adjust the strength of that lora as needed. in that lora, you would avoid tagging any hair or facial features (theorizing with this), so the two loras don't get confused, and the second primarily focuses on the glasses. That's the beauty of doing a lora of yourself, is that you are your own IP, so if you're willing to put in the time, you can diversify the hell outta the dataset and have it create a perfect 'you' every time. Now, setting up the camera, if you're using a smartphone and don't want the 'selfie' pose every time, is a different challenge in itself. Edit: sorry, It's 3am and I was only half paying attention. I went back over part of the video and realized that you already mentioned the glasses thing.

That tip about the clothing is actually pretty smart. ChatGPT, and even a search I did online suggested that if I didn't want the model to be trained on the clothing, and only wanted it to focus on the character, not to add the clothing tags, which made sense to me, but what you've just said actually lines up well with what I've experienced in image generation when stuff shows up that I don't want in there.

Today you need at least 3.10.9 , I found that out the hard way!

how add this in krita? thank you Bro

thank you Bro. i use adblock and tempermonkey translater.😎

Is it possible to add a custom diferent folder for checkpoints? Cause my main HD where Krita is installed is almost full. I have other models in a differente directory, but i cant figure out, how to add it. Tested everything and didnt work yet :(

You're incredibly articulated, your presence and voice are more powerful than the vast majority of people, myself included. It sucks to hear you had to contend with that violent drunk uncle, but clearly you are a warrior, whatever the story is it looks like he lost in my eyes. Good video, I am trying to start up a business and get out of my rut of where I am at in life and absolutely need to be able to expand images. I have hand drawn them on GIMP before but this can take almost a whole day for something which is a small detail in my overall work. Thank you for enabling me to do better

Thank you for this video. I just installed it into Krita, and it frankly is a freakin' marvel! I have a 12Gb 3090 Nvidia card, and a 1024x1024 image takes about 10-15 seconds to generate, so it's REALLY fast. The UI is actually richer and has more possibilities than what Photoshop offers for money, and the quality of the generation is miles ahead of Adobe's. Mind blown!

Hey man, I'm running Forge UI which requires me to have Python 3.10.6, but Koyha SS requires Python 3.10.11 to run, and it errors out. Do you remember which version of Kohya you've used for your video those 9 months ago? Maybe older version would be compatibile with it for me. Is there any way around it?

Just want to chime in and thank you for the json file with your settings. They worked wonders. VRAM usage perfectly fine for my 12GB RTX3060, and finished training fast with about 28-30 images. The resulting LoRA is very flexible and works a treat!

What kind of iteration time should one expect ? I use a 3090 and 100 images at a time

Great video! I still don´t understand how to use IP adapters on Krita. Could you do a tutorial on that?

13:10 what does "adding your own models" mean?

Just what I needed to know - thanks

its not really what im loking for :( Microsoft are killing paint3D for no reason and im looking for something like the paint3d magical select feature. it lets you *select any object in the picture, it then POPS it out as a movable separate object and then fills the background with some AI to make it look like there was nothing there.... * Its literally the best thing microsoft made and i dont know why they are getting rid of it next month 😭😭 but i cant find a program that does this feature :( please does anyone have any ideas thank you 😭😭🙏🙏

I get Cuda out of memory regardless of what settings I use. I also have a 3060Ti 8GB. What can I do?

Hey Streamtabulous, I wanted to ask: Usually there is an "output" folder for generated images using stable diffusion, that keeps on taking memory over time, is that the case with this krita plugin? coz I would like my generated images to be only taking space the ones I use not all of them that were generated but not used.

Hi there, could you please do one for GPUs above 470x , I'm rocking a 570 with 8gb but still I think its crashing in video memory or something, would be helpful to know.

All love and support ❤❤

You said a lot about the problems at the edges. Naturally, this happens when you set the Feather option to 1 at the beginning of your video. You should set it to the maximum value so that the edges soften and merge with the background.

Why so high LoRA alpha?

With 1.5 the settings would remain the same?

hey do you have best res to use for genai /stable diff, then later upscale? 500x500 px?

hey stream, thanks for this, im trying to add scifi yet realism to my photos, like a space ship in sky in my selfie or sand dunes on ground etc, what do you think is the best model with the settings?

so glad i found you, do you have a basic krita setup with different models video? and how to look for models to inject into krita etc.

Hi, i hope you can help me. I am trying this for some time and doesnt work for me: I add a new layer with a character on it, as you did. So i select it, add a new control layer, and add a pose control. Click on 3 starts, and start to calculate the skeleton. When its read. I got a new black layer, with the skeleton on it. I tried every step slowly and doesnt work. Maybe its a bug?

hey thank you for this! I wanted to ask how do we add different models for landscape models / car models etc..

Thank you for this. <3

thanks for the video!, has anyone had a feeling that the lore does not work, and the pictures are all the same during sampling?

When you tag, tag ONLY what the AI should ignore. Don't tag specific facial features of your face. Because you want yourself to be appearing consisten when generating images. Tag stuff that also changes like clothes, backgrounds, etc.

They just added support for custom ComfyUI nodes in the Krita AI Diffusion today. I would pay actual money (not much, I'm broke lol) for you to come back and do a tutorial on it.

Hi, I’m new to using krita generative ai plugin: I keep getting an error message saying I don’t have enough tokens is there a way to buy more?

Thank you for this tutorial, I've tried training SDXL on Kohya and the program crashed every time. I have tried this method with Pony and found that it does work, but the loras I make have to be at high weight to work but that could be because I didn't format the datasets properly. But one issue I have is that it takes 10 hours for me to finish training one lora. I have a Nvidia 4060 gpu which is an 8gb one and I'm not sure if there's something wrong with it (since I have to reset it occasionally or else SDXL will take 10 minutes to generate a single image) or if I'm just using the wrong hardware.

Glad i watched this. Its on sale for 20 bucks right now. I was trying to find a budget webcam for my series x. Saved the hassle.

i dont have setup.bat file. evrethyng was installed fine..but...where is the setup file ?

Wow, es el unico tutlrial que encontre de este tipo con Krita, lastima que me cuesta trabajo seguirte ya que mi ingles no es muy bueno, pero el resultado es hermoso 👍🏾👍🏾👍🏾

I know you have Nvidia graphic but could you by any means do some tutorial how to run the Krita AI on AMD graphics in Win11? So far I was unable to make it work on my new laptop which has AMD graphic (RX 6650M) and its such a shame :( I would be happy for any hint...

Video title says part 2 but I can't find a part 1 in your videos

Hi Steamtabulous, Do you have any idea on how to use two checkpoints at the same time here on Krita? Like one for generation and the second as a refiner?

I tried to create simple animation with "full animation" of a single ball bouncing in 15 frames but it did not work. Only dot color on a ball change. I don't know how but there is no guide about this.

5.2.6 works !! 10/4/2024