I have RTX 3090 on windows
I have RTX 3090 on windows
�I think you can use fp16 model but will be a bit slow
MI'm already downloading, so we'll soon find out
MThe FP16 model is no better, there is a problem with the generation. And with UniPC Normal I got only noise.
�In the data you sent I see you are using Euler sampler and simple schedule
MHere is UniPC result 
�Not sure what's happening, I use swarm UI and use these settings recommend by Dr. Furkan and it works for me
MI'm on the trail. I've generated a new config file for Forge, so I just need to compare the difference between the two to see which parameter is spoiling the image.
MWe have the cause of the problem! Setting the sigma churn to 1 caused the poor image quality. Previously with the other SD models this was not a problem, but the Flux doesn't seem to like it.
MTo conclude, here is the difference between the FP16 and NF4 models. I keep them both because of the difference, because sometimes I need a different generation, but for FP16 I prefer to use the GGUF version, I have more resources.
MDo you train T5-XXL by checking the Train T5-XXL box in Kohya?
MIs there anything else you need to do?
HI followed your runpod forge setup tutorial instead of using a1111, when I run the startup command I get this,
"Cannot locate TCMalloc. Do you have tcmalloc or google-perftool installed on your system? (improves CPU memory usage)",
does this effect anything? seems to work ok though. also runs out of gpu very fast on RTX 3090 but should be normal?
"Cannot locate TCMalloc. Do you have tcmalloc or google-perftool installed on your system? (improves CPU memory usage)",
does this effect anything? seems to work ok though. also runs out of gpu very fast on RTX 3090 but should be normal?
SI managed to get flux dev running on swarmui. But when I type: "portrait of maratwtf" it draws portraits of beautiful women instead of my ugly face lol
Sthe lora didn't connect to flux or something
DIs the 4 GPU training thing fixed?
DElse I shall use 1 GPU to avoid these issues.
GAnyone got a good ComfyUI workflow for trying out these new FLUX LoRA models? The X-Labs workflow is giving me an "list index out of range" error.
Nlatest update he posted mentioned he fixed it
DGuys, I didn't get 4 available GPUs, so I am using 2, it says in the vid:
New learning rate = (no. of GPUs * batch size / 2) * LR
= (2 * 1/2)* 0.00005
= 0.00005
For me, the formula just cancels out. Wth?
New learning rate = (no. of GPUs * batch size / 2) * LR
= (2 * 1/2)* 0.00005
= 0.00005
For me, the formula just cancels out. Wth?
Sguys where do I get epic prompts for my photos? Like the ones in ohmwx man videos
DCheck posts at civitai or ask chatgpt
DTbh, use swarmUI (even if you just use it for the grid generator feature; well explained by the Doc in the video). It just makes it soo much easier. But if you still want to use comfyUI, I think I saw a couple on civitai.
Fin which ui? iPNDM
Fah i see it is on swarmui
Fit is it new i had tested all
DYes, SwarmUI, it is good.
Fcurrently generating 1080 images
Ffor style testing
F
Fsamplers not working on forge
Flike unipc
Fworks on swarmui
Fthey added but i havent tested yet
Fnope doesnt impact
Fdid you enable lora
F4x gpu works. i trained straight 40 hours
Fbut he tries to train like 10000 images i dont know if that is broken
Fye dont change LR 
For maybe 0.00075
Dlol, I remember setting it back to 0.00005, but it defaulted to 0.0001 for some reason.
DI'm wondering whether that will break my training 
DBoth the GPUs are running fine though
Fnope should work fine
Fnice
XThis is my task manager before training
DEven close the task manager and google chrome windows. Just use nvitop to monitor.
XWow okay
Xthanks
Xwondering if Rank_12_7500MB_7_68_Second_IT.json will produce lower training givingit a shot now