do you see improved results?
do you see improved results?
Fi am about to compare 800 images generation
Fwaiting grid to finish
F780 - each one 48 steps FLUX image
Dyes but is a preliminary test, I stopped and restarted the training with different settings a couple of times
Dnow I started the real test
DI will report in a few hours
Dremember my goal is to train multiple people no concept or class bleading
Dlest see if I succeed
Fgreat
DIn my last test I wasn't using unique tokens and I changed tokens and removed class in the middle of a training resuming..
DIt was a mess
Dother note, all my regularization are captioned
D27hs left
Fnice. is this single gpu?
Drtx 3090
Dmultiple gpu can be used on kohya?
Fok it is really good speed
Fof course
Fi recorded tutorial
Fyou can even use 8 gpus :d
Fediting the tutorial
DI have just one
DF
Fyou can rent on massed compute
F1.25$ per hour 4x A6000 GPU
Fok i am gonna update configs with clip l
Fthere is slight improvement
Fi will make it seperate folder
Dhow it works you increment the batch size, it proceses 1 step per gpu or what?
Dyes I know but I'm just doing it for fun on my own computer
Fexplained in video please wait it :d
Dsure thanks
UHey guys, how can I use .gguf format models in SwarmUI?
UWarning: This is the 16-step model. The faster 8-step model can be found HERE [Note: Unzip the download to get the GGUF. Civit doesn't support it n...
Fi am gonna ask this now
UI just figured it out. For those who want to try out: In order to work, you need to manually edit metadata of the model and select architecture type. All gguf models goes into unet folder and after restart it automatically downloads Comfy GGUF Workflow.
UI'm testing this right now but so far it looks great for 12-16 gb VRAM GPU's.
DIs there like a ComfyUI XY Plot workflow to test/ compare multiple Flux Lora models?
Fwhy not use swarm?
DWould it mean relearning a lot of stuff?
Fnope
Fi shown how to make
FYouTubeSECourses
Ultimate Kohya GUI FLUX LoRA training tutorial. This tutorial is product of non-stop 9 days research and training. I have trained over 73 FLUX LoRA models and analyzed all to prepare this tutorial video. The research still going on and hopefully the results will be significantly improved and latest configs and findings will be shared. Please wat...
F52:55 Grid Generator tool for selecting LoRAs and prompts
53:59 Generating the grid and expected results
56:57 Analyzing grid results in Swarm UI
53:59 Generating the grid and expected results
56:57 Analyzing grid results in Swarm UI
Fway easier than comfyui
Fand uses comfyui at back end
DYea, I saw the video. Need to setup swarm first, then get it up and running
LHelp me to understand this error 
LI'm using the flux1CompactCLIPAnd_Flux1DevFp8 training model