Ive been tinkering with AI - mostly Stable Diffusion and learning to use Local LM- Been enjoying making and editing pics with AI for the most part. Gemini has helped me a ton on fixing my programs and optimizing a lot - I feel good about learning more.
Here are some examples of what I can do - Taking Pantas Comic pic from a couple of threads ago -
For comparison purposes - Google Nano Banana - 245 seconds - Clean - Not local but its clean - took 2 tries
5090 - Comfyui - Qwen - 4 step - 40secs - 1try
5090 - Comfyui - Qwen - 20step - 2mins - 1try
(Really can't see the difference outside of ink)
"But Nickeny, Nanobana beat you bro"
SD Upscale 4x - 1min
Yes, it lost some clothing detail, but considering I'm just some scrub learning from YouTube videos and asking AI for help, I think I am moving in the right direction.
I learned Loras and how to make them ,Winanimate,Wan2.2, infinitetalk too - and now got RFC Voice now too - Its what in unison this can do -I mostly use forge and Comfyui now
Thanks for coming
Bonus -
Created a Nickeny Lora file - Now he is the Cute *** NTRing Sirical in front of panta
I should make FFXI comics...