Software Engineering Courses (SECourses)•10mo ago

Would love to hear about it if you find a custom GPT or LLM Instructions that reliably understand ho

Would love to hear about it if you find a custom GPT or LLM Instructions that reliably understand how Flux likes to be prompted. Often, It feels like they just shotgun a cloud of text at the Clips.

SSegundus Would love to hear about it if you find a custom GPT or LLM Instructions that re...

AiInfluence•3/9/25, 12:04 AM

I mean the answer lies on the dataset of both the T5 and Clip models . If we had access to those and train an LLM to give prompts based on it it would be dope

SSegundus Would love to hear about it if you find a custom GPT or LLM Instructions that re...

Furkan Gözükara SECourses•3/9/25, 12:15 AM

joy caption really good

Furkan Gözükara SECourses•3/9/25, 12:15 AM

or cogvlm 2

Furkan Gözükara SECourses•3/9/25, 12:16 AM

we have both

humangirltotally•3/9/25, 2:42 AM

Can anyone maybe tell me why I have this weird noisey image? It's all models and loras

StayThePath•3/9/25, 2:58 AM

Anyone around her ehave some experience with training loras? I want to learn how to make them, but I'm not sure what is the most effective method of doing it and it seems like tehre are quite a few ways to do it. Anyone have a recommended way? I've been using comfyui pretty much exclusively for image gen and if quality is equally as good, I'd prefer comfy

SStayThePath Anyone around her ehave some experience with training loras? I want to learn how...

humangirltotally•3/9/25, 3:00 AM

onetrainer > kohya

StayThePath•3/9/25, 3:00 AM

onetrainer, I'll check that out. It's more the quality of lora I'm concerned about than anything else

StayThePath•3/9/25, 3:02 AM

any idea if onetrainer is planning to support wan?

blackhat•3/9/25, 7:54 AM

@Dr. Furkan Gözükara Im also training a style lora to get very realistic images. I have seen some LoRAs like these on civit ai https://civitai.com/models/652699/amateur-photography-flux-dev I'm trying to train to get this style but been struggling to get the skin texture right. I tried with 150 images, 8000 steps, batch size-2, rank- 64. I have also added 10 face closeups to get some skin texture. I mostly sourced images from a friend who is a photographer and then from stock image websites. What am I doing wrong? Is my dataset too less or im still picking clean looking images and should do more images which are actually taken from phone?

blackhat•3/9/25, 7:54 AM

Generations with my LoRA

blackhat•3/9/25, 7:54 AM

46a56669-69ec-47fb-8d26-e35b55ce141e_2469229373439378570_00002_.png

af816e00-df2b-47ef-a34e-494baeb0fdbd_8097623383787747964_00002_.png

blackhat•3/9/25, 7:54 AM

generations with the amateur photography lora

blackhat•3/9/25, 7:55 AM

Bblackhat @Dr. Furkan Gözükara Im also training a style lora to get very realistic images...

AiInfluence•3/9/25, 9:01 AM

128 Lora rank for max quality

Hhumangirltotally Can anyone maybe tell me why I have this weird noisey image? It's all models and...

Furkan Gözükara SECourses•3/9/25, 9:40 AM

probably inaccurate parameters

Furkan Gözükara SECourses•3/9/25, 9:40 AM

watch this video : https://youtu.be/-zOKhoO9a5s

YouTubeSECourses

Stable Diffusion 3.5 Large How To Use Tutorial With Best Configurat...

Stability AI published their most power newest model Stable Diffusion 3.5 Large. This model unlike FLUX is full model not distilled and has huge potential. I have done extensive research and publishing all of it in this video regarding how to use SD 3.5 Large with the best settings. Moreover, I am sharing how to use FLUX DEV with the best possib...

SStayThePath Anyone around her ehave some experience with training loras? I want to learn how...

Furkan Gözükara SECourses•3/9/25, 9:56 AM

i have step by step excellent tutorial with config

Furkan Gözükara SECourses•3/9/25, 9:56 AM

https://youtu.be/nySGu12Y05k

YouTubeSECourses

FLUX LoRA Training Simplified: From Zero to Hero with Kohya SS GUI ...

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...

SStayThePath any idea if onetrainer is planning to support wan?

Furkan Gözükara SECourses•3/9/25, 9:56 AM

nope not. for Wan training I plan kohya or ostris

Furkan Gözükara SECourses•3/9/25, 9:56 AM

probably kohya

StayThePath•3/9/25, 9:58 AM

Nice. I just spun up some runpod with a simple gradio ui.

SStayThePath Nice. I just spun up some runpod with a simple gradio ui.

Furkan Gözükara SECourses•3/9/25, 9:58 AM

i have runpod tutorial too

Furkan Gözükara SECourses•3/9/25, 9:58 AM

https://youtu.be/-uhL2nW7Ddw

YouTubeSECourses

Blazing Fast & Ultra Cheap FLUX LoRA Training on Massed Compute & R...

Unlock the power of FLUX LoRA training, even if you're short on GPUs or looking to boost speed and scale! This comprehensive guide takes you from novice to expert, showing you how to use Kohya GUI for creating top-notch FLUX LoRAs in the cloud. We'll cover everything: maximizing quality, optimizing speed, and finding the best deals. With our exc...

Bblackhat @Dr. Furkan Gözükara Im also training a style lora to get very realistic images...

Furkan Gözükara SECourses•3/9/25, 9:59 AM

your dataset

Furkan Gözükara SECourses•3/9/25, 9:59 AM

also you shouldnt increase batch size like that because batch size impact learning rate. you can use our lora config

MetaKan•3/9/25, 10:20 AM

hi @Dr. Furkan Gözükara could you tell me if this parameters of the SamplingFlux exists in SwarmUI? thnks

MMetaKan hi @Dr. Furkan Gözükara could you tell me if this parameters of the SamplingFlux...

Furkan Gözükara SECourses•3/9/25, 10:21 AM

not sure you can ask here

Furkan Gözükara SECourses•3/9/25, 10:21 AM

https://discord.gg/rz6Jw7ZM

Furkan Gözükara SECourses•3/9/25, 10:22 AM

coding practical rife batch processing app

Furkan Gözükara SECourses•3/9/25, 10:23 AM

increase FPS

Furkan Gözükara SECourses•3/9/25, 10:33 AM

i am gonna push limit of manus ai now

a very hard very comprehensive task i am giving

Furkan Gözükara SECourses•3/9/25, 10:55 AM

i am trying to add video upscale as well

Furkan Gözükara SECourses•3/9/25, 11:19 AM

prepared a 1114 words prompt for manus ai lets go

Mmasood how do you guys do that? first you take a base model and finetune it on iphone s...

julius•3/9/25, 11:39 AM

Yes that was my plan

TTimson Try 4:5 aspect ratio as it is dominant for mobile photos and Instagram. All you ...

julius•3/9/25, 11:43 AM

I trained on 1:1, is that still good?
In SwarmUI i choose 4:3 for generation, that is the native ratio of iPhone

AAiInfluence what token and class did you use?

julius•3/9/25, 11:44 AM

token i tried "jqxz" (didn't want to use ohwx bcuz i wanted to fine-tune the model again later on a person). Class i tried "style" but Furkan suggested to use "smartphone photo" next time

Jjulius token i tried "jqxz" (didn't want to use ohwx bcuz i wanted to fine-tune the mod...

AiInfluence•3/9/25, 11:53 AM

so you prompt like "a jqxz smartphone photo of ohwx man" ?

The90sPope•3/9/25, 12:10 PM

Guy is selling a 3090ti dirt cheap, I currently have 4080super.
I know 24gb vram is gonna fit a larger model, but what about speed?
Is it a decent card for AI video Gen or should I wait for a good deal on 4090.
5090 are impossible to get atm where I live.