Freally good
Fbut dont worth it i think
FA6000 gets 6.3 second
Fso it is like 2.4 times faster
Fbut how many times more expensive 
Ftry L40S
ïżœawesome, I will 
Yeah, I'm overpaying atm, but at least I can use my checkpoint tomorrow noon
Yeah, I'm overpaying atm, but at least I can use my checkpoint tomorrow noon ïżœL40S, got it, thanks!
X@Furkan GözĂŒkara SECourses Is it possible to put a new optimizer? like this one?https://github.com/derrian-distro/LoRA_Easy_Training_scripts_Backend/blob/main/custom_scheduler/LoraEasyCustomOptimizer/came.py
I saw few people use this optimizer and another thing called REX
I saw few people use this optimizer and another thing called REX
GitHub
The serverside backend created for use with the LoRA Easy Training Scripts Frontend - derrian-distro/LoRA_Easy_Training_scripts_Backend
Xlike this:- ss_optimizer": "LoraEasyCustomOptimizer.came.CAME
Xcreators who train characters use this, like to know how it different from others
S@Furkan GözĂŒkara SECourses doc, what's limiting my epochs to only 18 and 1600 steps? using the new 48gb configs in kohya.
Fyou can request from kohya only he can do
Fbmaltais error
Freported
Fi put a new bat file that switches back to older working commit
Fyou can use it
Fcd kohya_ss
git checkout d47e3e664286314c6b9728bb38b1c7a576211572
git checkout d47e3e664286314c6b9728bb38b1c7a576211572
R@Furkan GözĂŒkara SECourses one question for understanding. why would 16GB config be slower than 12gb config. isn't more VRAM better? I am comparing configs and only difference is blocks to swap. should it now slow down training if 12gb config has more blocks to swap ?
Rso for v14 it is necessary to use the bat file fix ? or is it only if this error is encountered
Ralso. 3rd question. do we still need Windows_update_kohya_and_fix_FLUX_Step2 ? or is it not necessary? @Furkan GözĂŒkara SECourses
SI used these steps in the terminal at kohya_ss folder, still the same issue
Feven kohya couldnt find it
Fit is a weird thing
Fyes but bmaltais fixed
Flatest version should work now
Fnot atm but in future maybe
Ftry latst version
Fmaybe now works
Fhe made an update
SSorry, what do you mean latest version? I'm V44 files from a day ago.
CL40S and A100 PCIe, are my first choices.
L40S is very polyvalent and has a good speed.
I use A100 PCIe when I don't want to loose time about searching for the parameters limits and fall into Out Of Memory problems.
L40S is very polyvalent and has a good speed.
I use A100 PCIe when I don't want to loose time about searching for the parameters limits and fall into Out Of Memory problems.
Fi mean run updater bat file
Fkohya gui fixed error
Si'm running ubuntu on massed
Fthen run installer
Fit will update
Four installer from zip file
Slink please doc
FFuse massed compute instructions
Sty ty doc!
SI'm creating Loras
Sit is the same?
Fyes
Fuses same files
Sok thank you again doc
Rthanks for the responses.
R@Furkan GözĂŒkara SECourses can we train FP8 models as well using same configs? or do I need to select fp8 in the training parameters. Atm I can train lora on FP8 using your fp8 config.
