Software Engineering Courses (SECourses)•5mo ago

I saw that there are cartoon - style characters in your videos.

筋筋斗云 I saw that there are cartoon - style characters in your videos.

Furkan Gözükara SECourses•8/17/25, 8:28 AM

yes they are hard

Furkan Gözükara SECourses•8/17/25, 8:29 AM

if you want cartoon flux kontext much better

Furkan Gözükara SECourses•8/17/25, 8:29 AM

https://youtu.be/adF9X9E0Chs

YouTubeSECourses

FLUX Kontext Dev Detailed Local Windows How To Tutorial - Better Th...

Black Forest Labs (BFL) publisher of FLUX models kept their promise and published the FLUX.1 Kontex DEV model today. So I didn't sleep and after doing huge research and test non-stop, I have prepared this excellent step by step tutorial that will show you how to use this amazing model. With FLUX Kontext you can edit any part of the image in any ...

Furkan Gözükara SECourses•8/17/25, 8:30 AM

and flux krea will make it even harder

Furkan Gözükara SECourses•8/17/25, 8:30 AM

for cartoon

筋

筋斗云OP•8/17/25, 8:31 AM

I can use kontext, and I have successfully trained my own control lora. At present, I just want to replicate the process in your tutorial, use a dataset of real - person photos to train a DreamBooth (DB) to achieve a personal model with generalization ability.

筋

筋斗云OP•8/17/25, 8:32 AM

Thank you for your reply.

筋筋斗云 I can use kontext, and I have successfully trained my own control lora. At prese...

Furkan Gözükara SECourses•8/17/25, 9:11 AM

use flux dev not krea for this task

Furkan Gözükara SECourses•8/17/25, 9:11 AM

flux krea is optimized for realism

筋

筋斗云OP•8/17/25, 9:31 AM

Thank you for your suggestion. I'm re - modifying the main model for multi - GPU retraining.

筋筋斗云 Thank you for your suggestion. I'm re - modifying the main model for multi - GPU...

Furkan Gözükara SECourses•8/17/25, 9:32 AM

for multi gpu loras works great with 48 gb gpus

Furkan Gözükara SECourses•8/17/25, 9:32 AM

dreambooth need 80 gb gpus

筋

筋斗云OP•8/17/25, 11:14 AM

4X 4090 48G

筋筋斗云 4X 4090 48G

Furkan Gözükara SECourses•8/17/25, 7:47 PM

awesome

Furkan Gözükara SECourses•8/17/25, 7:47 PM

i wish they were sold worldwide

Furkan Gözükara SECourses•8/17/25, 7:47 PM

definitely better than 32 gb 5090

Timson•8/17/25, 10:44 PM

Tried training character checkpoint with 50 image set 7000 steps from flux krea and result is much worse than flux dev counterpart so far. Much less likeness, seems more weird results

TTimson Tried training character checkpoint with 50 image set 7000 steps from flux krea ...

Furkan Gözükara SECourses•8/18/25, 12:31 AM

i feel krea need more epochs

TTimson Tried training character checkpoint with 50 image set 7000 steps from flux krea ...

K4bronsky•8/18/25, 1:00 AM

I found using a character Lora that I trained with flux on flux krea is giving me the textures of krea with the character. Vs the finetune I was getting fuzzy images

Tim0807•8/18/25, 1:05 AM

is a 80gb gpu needed for flux krea Dreambooth/Finetune?

As far as I remember for flux dev DB/Finetune a 48gb gpu was enough

TTim0807 is a 80gb gpu needed for flux krea Dreambooth/Finetune? As far as I remember f...

Furkan Gözükara SECourses•8/18/25, 1:18 AM

6 gb sufficient

Furkan Gözükara SECourses•8/18/25, 1:18 AM

if you want multiple gpu yes

zadkhamaya•8/18/25, 7:54 AM

Hi,

I was trying to generate using my trained dreambooth model on Runpod. It shown this error in SwarmUI. When I run it in massedcompute, it's working fine. Do you what is the cause of this error?

Thank you

Zzadkhamaya Hi, I was trying to generate using my trained dreambooth model on Runpod. It sh...

Furkan Gözükara SECourses•8/18/25, 11:49 AM

model download failed

Furkan Gözükara SECourses•8/18/25, 11:49 AM

rm -r modelpath

Furkan Gözükara SECourses•8/18/25, 11:49 AM

redownload

skyrrr•8/18/25, 1:58 PM

@Dr. Furkan Gözükara Hey Dr, is there a theoratical limit with checkpoint training regarding dataset ? I have a good 350 images, is that too much or I would just need to train it longer ?

uisato•8/18/25, 3:22 PM

https://www.instagram.com/p/DNQpj1rRPEV/?hl=es

uisato_

⟆ - tʀɪße_∞ : [2]

regⱭlia∴CΥΛN∴bøne⟲loop∴ritual

ƒ͓*c͓κ͓ †͓h͓ɘ͓ ͓α͓l͓ĝ͓ø͓ // attn=∅ ✱

#animation #cinema #design #tribal

Likes

371

Instagram

Uuisato https://www.instagram.com/p/DNQpj1rRPEV/?hl=es

Sakami•8/18/25, 3:35 PM

wan2.2?

Elli_Oaken•8/18/25, 6:36 PM

Hi Furkan, I want to confirm something about your captioning theory.
For a small and very consistent dataset (around 50 solid product photos on white background), is it still best practice to only use the token + class as caption (like CCCSNOOO bag), and let the images carry all visual attributes?
Or in such small homogeneous datasets, could adding more detailed captions (e.g. color, material, shape) actually help stabilize training with T5?

JonkoXL•8/18/25, 6:52 PM

Good question! AFAIk depends on what you want. iI you dont caption a certain thing, it will always generate it in that thing, unless you really force it in a prompt.
If you caption something (like color, material) you can more easily change the color , material later.
I think Furkan will confirm this

JJonkoXL Good question! AFAIk depends on what you want. iI you dont caption a certain thi...

Elli_Oaken•8/18/25, 7:25 PM

Thanks for the clarification! If T5 is active with such a small and homogeneous dataset, could relying only on token + class captions cause it to fall back to it's pretraining bias (e.g. anime/fantasy)? In that case, would the safer option be to disable T5?

Also, around what epoch should the results typically start to become visible? I’ve tested the token + class approach, but even after 190 epochs I still mostly get strange anime-like outputs. Does that indicate that the LoRA simply needs more training, or that the captioning strategy itself is unsuitable for this setup?

JonkoXL•8/18/25, 7:27 PM

I have no idea about T5, is that Clip?
What model are you training on? SDXL, Pony , Flux, Qwen, Wan 2.1, Wan 2.2?

JonkoXL•8/18/25, 7:27 PM

What's your learning rate

JJonkoXL What's your learning rate

Elli_Oaken•8/18/25, 7:29 PM

Flux. It's this config setup.

24GB_GPU_Quality_Tier_1_22950MB_14.1_Second_IT_Trains_T5_and_T5_Attention.json5.6KB

JonkoXL•8/18/25, 7:29 PM

Are you able to generate images while training?

JJonkoXL Are you able to generate images while training?

Elli_Oaken•8/18/25, 7:30 PM

No, but I have three computers.

JonkoXL•8/18/25, 7:30 PM

I think your learning rate is extremely low

JonkoXL•8/18/25, 7:30 PM

now I havent trained Flux, but 5e-05 is pretty low

JonkoXL•8/18/25, 7:30 PM

meaning 0.00005

JonkoXL•8/18/25, 7:31 PM

try learning rate 2e-4

JJonkoXL now I havent trained Flux, but 5e-05 is pretty low

Elli_Oaken•8/18/25, 7:31 PM

Yes, 5e-5 looks low compared to typical SDXL LoRAs. But for Flux1 it’s the standard recommendation (e.g. Furkan’s config). It trains slower, but keeps things stable since Flux is more sensitive than SDXL/1.5. But maybe I'm wrong.

JonkoXL•8/18/25, 7:32 PM

Yeah I am coming mostly from SD 1.5 , SDXL and Wan

JonkoXL•8/18/25, 7:32 PM

i still would try going for 2e-4 with flux and see if you get decent results around 1500-3000 steps

JonkoXL•8/18/25, 7:32 PM

Worth a shot if you are not happy with your results

JonkoXL•8/18/25, 7:33 PM

try to find out if you can generate images while it's training (that means , the training algorithm stops for a moment to generate an image based on a prompt set)

JonkoXL•8/18/25, 7:33 PM

so every 100 steps generate two images, one with prompt identical to a training image, and another one with something you may want to prompt later

JonkoXL•8/18/25, 7:33 PM

the later is more important

JonkoXL•8/18/25, 7:36 PM

Yes , samples

JJonkoXL Yes , samples 🙂

Elli_Oaken•8/18/25, 7:42 PM

Thank you very much!

I saw that there are cartoon - style characters in your videos.

Similar Threads