Hello everyone. I am Dr. Furkan Gözükara. PhD Computer Engineer. SECourses is a dedicated YouTube channel for the following topics : Tech, AI, News, Science, Robotics, Singularity, ComfyUI, SwarmUI, ML, Artificial Intelligence, Humanoid Robots, Wan 2.2, FLUX, Krea, Qwen Image, VLMs, Stable Diffusion
@Dr. Furkan Gözükara I already have the training, but I have 99 @LORAS, is there a quick way to get the same prompt for all 99 @LORAS to see which one I get?
Hello dear @Dr. Furkan Gözükara I hope you are very well, I already have my LORAS in huggingface, I need to select the best one and then look for an economic tool to raise it via API... which one do you recommend? it will make about 100-200 calls per day. thank you very much, I have learned a lot thanks to you!
regarding details, which is better in your opinion SDXL Dreambooth (which is almost full finetuning) vs FLUX Kohya LoRA (which is NOT full finetune) for faces at the moment?
I am getting below error on 4 GPU, Using "4x_GPU_Rank_1_SLOW_Better_Quality.json" with just changing image directory and class prompts under data set preparation Note : (It was working in single GPU) @Dr. Furkan Gözükara
Is there something wrong with my settings? Why is the image I created so blurry? I use 5. Download FLUX.1-dev FP16 and 8. Download T5 XXL text encoder FP16 version
Flux has a very big problem with white backgrounds. Try not to use word Background at all. Try same prompt with red instead of white and it will not be blurry. Also lower guidance to 2 that will fix 50% of the blur. Also try diferent samplers
M
F
D
�
B
J
S
P
P
