you have to reduce resolution for speed up but reduces quality
you have to reduce resolution for speed up but reduces quality
Fyou can do 768x768
DThanks. Lol i thought i was way off. 25 seems fine
M@Dr. Furkan Gözükara doesn't seem like comfy or forge accept loras trained with text encoder yet
Ftext encoder training didnt added to kohya yet
Fso i dont know :d
Mi got it to work 
in sample images its fine, but they don't work on comfy or forge yet
in sample images its fine, but they don't work on comfy or forge yetFwell i am waiting kohya
Fhe said he found today gonna add :d
DFor multi gpu training do the gpus need to be equal in vram? Like i have an 8 gb card im not using and a 12 gb card that is installed. Would the 8 gb card work for multi gpu training with the 12 gb one?
Fvery likely
Fit loads onto each gpu cloning basically
DLooks like one trainer might be getting closer to flux release. Their dev branch is now called flux. So fingers crossed
DThanks
Fanyone know how to add a lora on comfyUI for flux dev?
Four configs ready :d
Fanyone know how to add a lora on comfyUI for flux dev? This is the workflow
DYou need to add a lora loader after the clip. Model from flux loader to left model port of lora loader. Clip out to lora loader and then model out to scheduler i believe
DI have a basic workflow that i can post that shows it
DOr post the image
DThis workflow allows you to harness the power of FLUX.1 [dev] within ComfyUI. Follow these steps to set up and run the model. The true "simplest" v...
D
DHopefully discord doesnt strip the data from the png
DIf it does thats the civ link to the json and workflow
DIt works for me and im noodle programming impaired so anyone should be able to use it
Fdiscord doesnt strip meta it works
F8 more trainings started
FThanks a lot, I placed t5xxl_fp16.safetensors in ComfyUI/models/clip/ folder but I can't find it here
DIts on the dual clip loader below that
DIf you want fp16 weights use default on that loader in tje pic
�@Dr. Furkan Gözükara Any clear way to make a grid table that's 4 columns across and 10 rows down for images using the Grid Generator tool in SwarmUI? I have 40 loras I wish to test one prompt and seed on but there's no clear instructions. The x/y/z plot in A1111 was so much more straightforward for this basic function 
Fdid you watch my grid part?
FYouTubeSECourses
Do not skip any part of this tutorial to master how to use Stable Diffusion 3 (SD3) with the most advanced generative AI open source APP SwarmUI. Automatic1111 SD Web UI or Fooocus are not supporting the #SD3 yet. Therefore, I am starting to make tutorials for SwarmUI as well. #StableSwarmUI is officially developed by the StabilityAI and your m...
Fput first tab loras
Fsecond tab prompts
Fthat is what i do
�I get it that this one is more powerful and advanced but holy shit it took me an hour just to get something close enough to what I was looking for. At first I just selected first field for Loras and autofilled that section and then it said it was generating a million images lol
�Yes I watched that video but you were showing the different upscalers there. I'll look again
Fjust do small test
Flike 2 lora
F2 prompt
Fand see results
Fthat will give you idea
Falso play with very top options
�Ok thank you @Dr. Furkan Gözükara I think I'm figuring out now. At first I was just using prompt settings from top left only but now I have the prompt on second tab and it's getting much more closer to what I was going for
�Sorry I get so easily frustrated, I wanted to figure it out on my own and not have to bother asking you until I had no other option. Very much appreciated, Dr! 
Fi got to sleep hoepfully see you tomorrow
�Goodnight Dr. I just ended up using this workflow instead in the Comfy tab: https://civitai.com/models/652483/flux-lora-tester
Hello 
for whatever reason you want to show me appreciation, you can: 
Ko-Fi With this workflow you just save time. A lot of time. Have fun. To...
for whatever reason you want to show me appreciation, you can: Ko-Fi With this workflow you just save time. A lot of time. Have fun. To...
�I completely had forgotten how much VRAM it was going to take to load up 40 Loras at once 
but this works good for now
but this works good for now