that is the whole world difference
that is the whole world difference
ZWell, to create serious works, the consistency of the character and the uniformity of the style are crucial.
ZLet’s focus on how this issue can be resolved?
ZIs it possible, for example, to blend a style or object Lora into your custom model? (I’m not talking about another character or a man.)
Fnormally a good trained style should work
Fbut civitai loras or models are all overfit
Fhow can you solve is a good research
Fmost likely training style and person together
ZHow to reduce the size of an extracted LoRA from 11GB to normal size around 600MB + - more less ?
Fextracted lora loses quality when you reduce the lora network rank
Fonly way is that
Falso save as fp8
ZI performed the extraction in FP8, and the size remained the same at 12GB.
ZThat is, from the FP8 model, which became 12GB, I performed LoRA extraction, and the size did not change at all.
Zsame 12 gb FP8 to 12 gb Lora...
Fwell that is just impossib le you made mistake somewhere
Fbest lora with fp8 is like 2-3 gb
Zwhats wrong here? in the step 2 i choose my finetuned FP8 model
Fsave precision
Fmake it fp8
ZAlright will try, it's just that this isn't mentioned anywhere, or did I miss it??
Fit is for saving space not mandatory
Ffloat is best quality
ZThere is no option to select FP8... only FP16, BF16, Float, and None.
ZThe 11GB size of LoRA… that’s why it’s better in quality. There’s no magic here. If LoRA weighed 1GB while maintaining quality, that would be a real breakthrough. But models with a size of 23GB, where LoRA doesn’t work properly, and extracted LoRAs at 11GB, are a significant negative effect that needs to be known in advance.
Fok then you need 2 steps
ZWhich steps?
Fconvert to fp8
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
ZIt reduced the size, but the image is completely black.
Z
Z@Furkan Gözükara SECourses This doesn’t work. I performed LoRA extraction from the model with conversion parameters set to FP16 quality, then converted this LoRA to FP8, and again got a black screen.
ZAre you sure your method works?
Fyep working
Flooks like you did accurate maybe something went wrong or swarmui broken etc
NI know there are much better ways to train a face, but I've got some credits I need to burn through on dreamlook.ai - can anyone suggest best settings to use with their expert mode?
Qfor a multi-gpu system Kohya cannot train. It gives me an error for gpu id not set. I tried to specify the GPU id in the Windows_start_Kohya_SS.bat file by somthing like this
set CUDA_VISIBLE_DEVICES=5
changing the device IDs but it didnt start -
running from the Install.bat menu worked for me but it is using the first GPU - it would be great to run paralell if they cannot run across the GPUs
set CUDA_VISIBLE_DEVICES=5
changing the device IDs but it didnt start -
running from the Install.bat menu worked for me but it is using the first GPU - it would be great to run paralell if they cannot run across the GPUs
�@Furkan Gözükara SECourses can a rank 256 fp16 lora work well if I'm using the fp8 version of my flux dev fine tune? or does the lora have to be fp8 as well?
Fas far as i know multi gpu jnot working on windows
Fbut if youy want to train on different gpuıs
Fuse accelerate gpu id setting
Fit should work
TBlack screen for me also after conversion to lora fp8
�in Acceleration you can set GPU, and multi GPU, but for fine tune using multi GPU you need 80 GB GPUs, if I remember correctly. No need to edit the bat file
�but it's better to convert it to fp8? 
Fnot sure. i keep loras as fp32 and base models fp16 
Fswarmui is able to auto cast into used precision
�you must have a lot of diskspace
FHi Doctor ! @Furkan Gözükara SECourses Please.. do you have some tutorial about training sd3.5? I prefer it to flux for some kind of generations.. I feel it more creative when subjects are not real humans ....
Fthank you
