Software Engineering Courses (SECourses)•9mo ago

I rented massed compute first time after few month off since i have rtx 4090... I though generating

I rented massed compute first time after few month off since i have rtx 4090... I though generating wan2.1 will be much faster....lol.. definitely not on L40S lol slow.. Maybe H100, but to expensive..

Ccyberbol I rented massed compute first time after few month off since i have rtx 4090... ...

Furkan Gözükara SECourses•3/31/25, 9:01 PM

true. 4090 is still a beast

Furkan Gözükara SECourses•3/31/25, 9:01 PM

and china selling 48gb and 96 gb . 48 gb is literally l40s

Kevin•4/1/25, 4:41 AM

man I'm trying to do dreamboothsdxl training to the likeness of an individual and I've tried so many times and the sample output images continue sucking. I subscribed to the Patreon, can anyone point me towards the most useful pages re: optimal parameters and so on?

Kevin•4/1/25, 5:08 AM

If your training data set is 59 images and each is being repeated 10 times, how many regularization images is ideal? Also, is repeating 10 times the best number of repeats?

KKevin If your training data set is 59 images and each is being repeated 10 times, how ...

Furkan Gözükara SECourses•4/1/25, 10:10 PM

hello sorry for late reply

Furkan Gözükara SECourses•4/1/25, 10:10 PM

i recommend you to train flux and we dont use reg images there

Furkan Gözükara SECourses•4/1/25, 10:10 PM

here best tutorial

Furkan Gözükara SECourses•4/1/25, 10:11 PM

https://youtu.be/FvpWy1x5etM

YouTubeSECourses

FLUX Full Fine-Tuning / DreamBooth Training Master Tutorial for Win...

If you want to train FLUX with maximum possible quality, this is the tutorial looking for. In this comprehensive tutorial, you will learn how to install Kohya GUI and use it to fully Fine-Tune / DreamBooth FLUX model. After that how to use SwarmUI to compare generated checkpoints / models and find the very best one to generate most amazing image...

Kevin•4/1/25, 10:12 PM

Thanks apprecaiate it, I have since switrched to flux and am working on that currently.

Kevin•4/1/25, 10:12 PM

Would you say FLUX is the current SOTA for photoealistic images trained to a particular individual?

KKevin Would you say FLUX is the current SOTA for photoealistic images trained to a par...

Furkan Gözükara SECourses•4/1/25, 10:14 PM

100%

Kevin•4/1/25, 10:15 PM

I bought the personal consultation thing at Patreon, how do we set that up?

KKevin I bought the personal consultation thing at Patreon, how do we set that up?

Furkan Gözükara SECourses•4/1/25, 10:17 PM

what is your username?

Furkan Gözükara SECourses•4/1/25, 10:18 PM

you can private message me here

Furkan Gözükara SECourses•4/1/25, 10:18 PM

ok i see it

Furkan Gözükara SECourses•4/1/25, 10:18 PM

message me here so we can setup. we can do now too

saberu0126•4/3/25, 4:19 AM

I did too- Gold membership

Ssaberu0126 I did too- Gold membership

Furkan Gözükara SECourses•4/3/25, 11:46 AM

hello please message me details . did you subscribe gold? private message me

moomoowowcow•4/3/25, 3:59 PM

Are there any special instructions or 'Additional parameters' to enter into the Kohya_ss GUI when trying to "Resume from saved training state (path to "last-state" state folder)?"

I have entered the path/workspace/stable-diffusion-webui-forge/models/Stable-diffusion/fruit_realismByStableYogi_v5XLFP16/model/ohwx-woman_Fruit_Full-finetune_adafactor_realismIllustriousBy_v35FP16_292-Img-200-Instances-step00032000-state

I have verified the contents within that folder include all of the expected, model_1, model, optimizer, random_states, scheduler.

The training starts, however, it starts at step 01, versus step 32001 as I intend to continue traning.

Can anyone offer insight or help here? Thank you in advance!

Q•4/3/25, 10:27 PM

does kohya work with rtx 5090?

QQ does kohya work with rtx 5090? 🤔

Furkan Gözükara SECourses•4/3/25, 11:00 PM

yes

Furkan Gözükara SECourses•4/3/25, 11:00 PM

we have updated our installers just use it

Mmoomoowowcow Are there any special instructions or 'Additional parameters' to enter into the ...

Furkan Gözükara SECourses•4/3/25, 11:01 PM

just use your last checkpoint

Furkan Gözükara SECourses•4/3/25, 11:01 PM

i never used last state

Furkan Gözükara SECourses•4/3/25, 11:01 PM

with our config exactly same

FFurkan Gözükara SECourses we have updated our installers just use it

Q•4/5/25, 8:13 AM

Thanks!

i've just upgraded my setups from 4090 to 5090 and noticing that I have problems running pretty much everything I tried before

thanks again! Now i think im getting it

QQ Thanks! 😄 i've just upgraded my setups from 4090 to 5090 and noticing that I h...

Furkan Gözükara SECourses•4/5/25, 10:45 PM

great

HRK•4/6/25, 5:56 AM

Hey, I wonder, has anyone tried / have article or any sort of content about full checkpoint finetuning on Wan ? I really want to get my hands into it but it'd be great to have some references to start working on it

HRK•4/6/25, 5:56 AM

as far as I know only DiffSynth offers the possibility to full checkpoint training on Wan, right ?

HHRK Hey, I wonder, has anyone tried / have article or any sort of content about full...

Furkan Gözükara SECourses•4/6/25, 12:41 PM

good point. i havent checked that yet though. currently musubi is good for lora

�

🍭🎀 𝒜𝓋𝒶 𝐹𝓇𝒾𝑔𝑔 🎀🍭•4/6/25, 2:19 PM

What are the optimal sdxl fine tune and lora configs? Having a hard time finding them. Using massedcompute. Onetrainer or dreambooth? Was thinking of using big love xl 2.5 as base checkpoint. Like the realism and quality of pics there.

�

🍭🎀 𝒜𝓋𝒶 𝐹𝓇𝒾𝑔𝑔 🎀🍭•4/6/25, 2:19 PM

@Furkan Gözükara SECourses

�🍭🎀 𝒜𝓋𝒶 𝐹𝓇𝒾𝑔𝑔 🎀🍭What are the optimal sdxl fine tune and lora configs? Having a hard time finding...

Furkan Gözükara SECourses•4/6/25, 3:34 PM

i have dreambooth config

Furkan Gözükara SECourses•4/6/25, 3:34 PM

but not for lora. for lora i recommend do dreambooth and extract

�

🍭🎀 𝒜𝓋𝒶 𝐹𝓇𝒾𝑔𝑔 🎀🍭•4/6/25, 6:16 PM

https://www.patreon.com/posts/very-best-for-of-89213064 this one?

Patreon

The Very Best Workflow For SDXL DreamBooth / Full Fine Tuning - Res...

Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon

�🍭🎀 𝒜𝓋𝒶 𝐹𝓇𝒾𝑔𝑔 🎀🍭https://www.patreon.com/posts/very-best-for-of-89213064 this one?

Furkan Gözükara SECourses•4/6/25, 7:01 PM

yes this is for kohya

Furkan Gözükara SECourses•4/6/25, 7:01 PM

this one onetrainer : https://www.patreon.com/posts/96028218

Patreon

OneTrainer Stable Diffusion XL (SDXL) Fine Tuning Best Presets | SE...

Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon

Mathew•4/7/25, 3:04 AM

I saw that in 24GB_GPU_Quality_Tier_4, it's configured with "blocks_to_swap": 0,
but in 24GB_GPU_Quality_Tier_1, it's set to "blocks_to_swap": 29.

Theoretically, wouldn’t using 29 blocks to swap reduce quality/consistency?
Or is it necessary to use it, otherwise there would be a VRAM impact?

Mathew•4/7/25, 3:05 AM

I'm just getting started with the more technical parts, so I might be saying something wrong if that's the case, I apologize.

JdotBdot•4/7/25, 11:43 AM

Hi everyone! Great to be here—thanks for all the amazing tutorials and shared resources.

I’m currently running into an issue with my first Flux LoRA training on Koya. I’m on an RTX 3080 (16GB VRAM) and followed the install/setup steps as advised. All required models are downloaded. Everything seems in place, but when I try to run the training, kohya starts up, seems to run ok then without warning goes black, crashes and restarts. It’s happened several times now.

Not sure what I’m missing—or doing wrong? any advice would be hugely appreciated. Thanks in advance!

MMathew I saw that in 24GB_GPU_Quality_Tier_4, it's configured with "blocks_to_swap": 0,...

Furkan Gözükara SECourses•4/7/25, 2:38 PM

block swap only changes vram usage - thus speed not quality

JJdotBdot Hi everyone! Great to be here—thanks for all the amazing tutorials and shared re...

Furkan Gözükara SECourses•4/7/25, 2:38 PM

hello

Furkan Gözükara SECourses•4/7/25, 2:38 PM

how much RAM you have?

Furkan Gözükara SECourses•4/7/25, 2:39 PM

this sounds like out of RAM error

FFurkan Gözükara SECourses block swap only changes vram usage - thus speed not quality

Mathew•4/7/25, 3:18 PM

Oh got it, thanks a lot for clearing that up!

MMathew Oh got it, thanks a lot for clearing that up!

Furkan Gözükara SECourses•4/7/25, 3:43 PM

you are welcome

FFurkan Gözükara SECourses you are welcome

Mathew•4/7/25, 4:53 PM

Oh, just one more thing. For example, is it possible to 'pause' a LoRA at checkpoint 50 and then resume from that point to continue training up to 100 or even 200?

MMathew Oh, just one more thing. For example, is it possible to 'pause' a LoRA at checkp...

Furkan Gözükara SECourses•4/7/25, 5:02 PM

it has use last lora weights

Furkan Gözükara SECourses•4/7/25, 5:02 PM

on interface you can try

Furkan Gözükara SECourses•4/7/25, 5:03 PM

tips-info

I rented massed compute first time after few month off since i have rtx 4090... I though generating

Similar Threads

Similar Threads

Similar Threads