Software Engineering Courses (SECourses)•5mo ago

until publish gui ?? ohhhh..

Ccyberbol until publish gui ?? ohhhh..

Furkan Gözükara SECourses•8/15/25, 1:57 PM

yep

Furkan Gözükara SECourses•8/15/25, 1:57 PM

i need to train and test

cyberbolOP•8/15/25, 1:58 PM

ok, so i will use just command by now

cyberbolOP•8/15/25, 2:05 PM

I'm looking for the latest info and best practices for training WAN 2.2 LoRAs, especially for character likeness on a 24GB card.

From what I've gathered so far, it seems the best dataset approach is to train the high-noise model on low-resolution videos (to learn motion) and the low-noise model on high-resolution images (to learn the fine details).

I also saw in the new Musubi-Tuner docs that you can apparently train both high and low noise LoRAs at the same time now, using the --dit (for low) and --dit_high_noise (for high) parameters in a single command. For 24GB VRAM, the key seems to be adding the --lazy_loading option to only keep the active model on the GPU.

Has anyone tried this simultaneous training method yet?

Ccyberbol I'm looking for the latest info and best practices for training WAN 2.2 LoRAs, e...

JonkoXL•8/15/25, 2:51 PM

No need for the lazy-loading as you can just train the two lora's after each other (in command line it auto continues to the 2nd command)

JJonkoXL No need for the lazy-loading as you can just train the two lora's after each oth...

cyberbolOP•8/15/25, 2:58 PM

I know. But y can also Train both same time

JonkoXL•8/15/25, 2:59 PM

It probably wont be faster, and you could decide to just turn off your pc after the 1st lora training if you decide that

cyberbolOP•8/15/25, 2:59 PM

Probably you right. I just experiment..we will know much better in few weeks probably..

JonkoXL•8/15/25, 2:59 PM

+ I may want to use less epochs or learning rate for the other model

JonkoXL•8/15/25, 3:00 PM

i noticed the high noise model gets overtrained quicker

cyberbolOP•8/15/25, 3:01 PM

@JonkoXL Do you successfully Train any high and low noise LORA and it's working together with OTHER lora keeps human eyes good quality?

JonkoXL•8/15/25, 3:04 PM

I think eyes in Wan 2.2. will still be a problem unless we get an even higher base resolution

JonkoXL•8/15/25, 3:04 PM

https://civitai.com/models/1862320/wan22-t2v-lora-cartoon-style

JonkoXL•8/15/25, 3:04 PM

This is my only Lora I succesfully trained so far

JonkoXL•8/15/25, 3:05 PM

haven't tried it with other loras

cyberbolOP•8/15/25, 3:06 PM

I Train three and when using with others LORA human eyes always problem.
Wan2 1 was always fine

JonkoXL•8/15/25, 3:07 PM

have you tried switching your high and low lora?

JonkoXL•8/15/25, 3:07 PM

for some reason i sometimes do that and apparently it can work out

JJonkoXL have you tried switching your high and low lora?

cyberbolOP•8/15/25, 3:07 PM

Oo. No..but will try when back home

JonkoXL•8/15/25, 3:08 PM

I played a bit with Wan 2.1 Lora training too, it seemed easier or more forgiving. But only tested it

JJonkoXL I played a bit with Wan 2.1 Lora training too, it seemed easier or more forgivin...

cyberbolOP•8/15/25, 3:11 PM

For wan2.1 I did some Loras

https://tensor.art/u/842260292657429385/models

CyberAImania's Models

JonkoXL•8/15/25, 3:15 PM

OOh nice :p

JonkoXL•8/15/25, 3:15 PM

what's your learning_rate and how many steps did you train for?

cyberbolOP•8/15/25, 3:18 PM

Wan2.1? I don't remember..but I Keep on drive all dataset and configs.

Ccyberbol 4090 (24 GB VRAM) – it was fine, but I’m now collecting a better dataset for my ...

Sakami•8/15/25, 3:19 PM

thanks - right now im doing Wan2.2 training on my dataset ll see how it turns out

SSakami thanks - right now im doing Wan2.2 training on my dataset ll see how it turns ou...

cyberbolOP•8/15/25, 3:20 PM

Let me know please. Share also what datasets you used if you can. Just photo's or mixed or video clips?

Sakami•8/15/25, 3:22 PM

due to nda i cant share dataset but these are studio photos and few natural shots of a person. Unfortunetly only photos for now

masterblaster3606•8/15/25, 5:28 PM

hello, i got out of memory error using massed-compute 2x RTX 6000 ADA

Batch_Size_7_48GB_GPU_46250MB_29.1_second_it_Tier_1_multi-GPU.json5.03KB

Mmasterblaster3606 hello, i got out of memory error using massed-compute 2x RTX 6000 ADA

JonkoXL•8/15/25, 5:34 PM

training what?

JJonkoXL training what?

masterblaster3606•8/15/25, 6:04 PM

40 images of a human for Flux

Sakami•8/15/25, 6:07 PM

No LoRA vs LoRA Wan2.2

Sakami•8/15/25, 6:08 PM

she is little bit damaged tho its 40 epoch [65img's dataset] 32DIM

JonkoXL•8/15/25, 7:49 PM

learning rate?

Mmasterblaster3606 hello, i got out of memory error using massed-compute 2x RTX 6000 ADA

Furkan Gözükara SECourses•8/15/25, 8:00 PM

did you load lora config into dreambooth or dreambooth into lora?

Furkan Gözükara SECourses•8/15/25, 8:00 PM

config name?

FFurkan Gözükara SECourses did you load lora config into dreambooth or dreambooth into lora?

masterblaster3606•8/15/25, 8:07 PM

No, I loaded dreambooth config into Dreambooth. I tried both 48GB_GPU_28200MB_6.3_second_it_Tier_1 and Batch_Size_7_48GB_GPU_46250MB_29.1_second_it_Tier_1. out of memory errors for both

Mmasterblaster3606 No, I loaded dreambooth config into Dreambooth. I tried both 48GB_GPU_28200MB_6....

Furkan Gözükara SECourses•8/15/25, 8:25 PM

what is your images resolution?

Furkan Gözükara SECourses•8/15/25, 8:25 PM

i recommend restart machine and use this config load into dreambooth

Furkan Gözükara SECourses•8/15/25, 8:25 PM

will work 100%

Furkan Gözükara SECourses•8/15/25, 8:25 PM

48GB_GPU_28200MB_6.3_second_it_Tier_1.json

JJonkoXL learning rate?

Sakami•8/15/25, 8:52 PM

3e-4