UPDATE: I completely forgot that Civitai provides SDXL LoRA training (my datasets are trained with LoRA, and then merged into the model).
You can support me by donating your Buzz to me for now.
===
Hi. As some of you might know, I started rolling out the first alpha version of AingDiffusion XL (ADXL) approximately 3 days ago.
I've begun training my dataset for the model, but I've encountered some issues. It took me almost 6 hours to train on my poor RTX 2060 Plus (12GB of VRAM). I noticed that the initial release lacked flexibility, forcing me to prepare a lot more datasets than I initially collected. I have three options for now. Either I have to stop AingDiffusion (SD 1.5) development completely to prioritize ADXL training, run the training on a cloud GPU, which is very expensive for a uni student like me, or maybe just give up.
My GPU is burning; HWMonitor shows that the GPU hot spot is at ~103 Celsius (~217 Fahrenheit). I'll probably wear out my GPU before finishing ADXL training at this point.
I know I will sound very selfish, but I need your help. I need some funding to at least rent a GPU on a cloud service. Any amount you send me through ko-fi (Here) will help me a lot.
Thank you in advance.
(If you want to know how much money I have right now when I write this; it's currently under $9 until the new year)