Where are your latest configs? Here?
Where are your latest configs? Here?
Fthis one has latest dreambooth and lora configs
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
NI saw a very advanced chtege that Visual Studio Code to it by api tied Sonnet 3.7 and it makes itself applications and sparate
Fcan you link
NIt was at a conference no record I don't have and no links but right before us she wrote a game with a working interface
NThis thing created the python file structure, then wrote a bunch of method classes, and the inerface separately with comments
NI'll share the links as soon as I get them
Bcan u mention of which article are you talking about in this
B@Dr. Furkan Gözükara
Fno article of this
Feach batch size requires new learning rate i explained in tutorial
Bokay
Bthanku so much
Fyou are welcome
Falso dont change network lora rank and lora alpha
Fthey impact learning rate too
Bas per your tutorial in the vedio right
B@Dr. Furkan Gözükara where is the vedio bcz in this artical i can't see that https://civitai.com/models/652699/amateur-photography-flux-dev
FYouTubeSECourses
Unlock the power of FLUX LoRA training, even if you're short on GPUs or looking to boost speed and scale! This comprehensive guide takes you from novice to expert, showing you how to use Kohya GUI for creating top-notch FLUX LoRAs in the cloud. We'll cover everything: maximizing quality, optimizing speed, and finding the best deals. With our exc...
Fthis one also has multi gpu training
HAnyone know how to get more realistic skin texture?
Ftraining with proper dataset
For use custom trained community models
HDoes this look unusual for settings?
Fchange your base model to community fine tuned ones
Dtook a while, but got it into a LoRA
D
HLooks like 360p image I have that issue too
Ajbox's Lora
Adiscord compression
Aoh and btw the grillz model is trained on plastic cartoon characters
HHow many steps are you using and what scheduler/sampler?
A50 ipndm and ddim
AI'm on swarm
Hnot comfy?
Ait's the same
AI'm using comfy as backend
Hddim uniform?
Ayes
Hlol 200 steps beta
H
Addim
Anot this
HThats my previous settings for the image, I am trying ddim and ipndm now! its taking a bit
Tflux guidance 1.7 very low
HI'm at 4.0 now and most my generations
Tfunny thing i worked with that exact latent upscale workflow today
Arule of thumb is your Lora must be rank 128 and trained on high quality realistic images
Afinetuned checkpoint even better but expensive
