As a photographer and AI enthusiast and technical artist I experimented everything possible long before midjourney and such came along (back in the days of discu diffusion in 2021). So I got a head start. Of course as a photographer I was very interested on how to use it for photography, and upscaling in this case. All commercial tools at this point suck for me.
But working with open source tools like comfyUI I finally managed to get something incredible. Example here : (first is 100% crop of tohe eye in the last image, second is upscaled version of the eye and then zoom out to original).
Now comfyUI is not a very easy thing to learn if you are not tech savy, but with proper searching you can also find or make yourself. If you want more info on the method I use, installing Comfy, needed files and links for this I will post a tutorial on my Patreon next month (I am now outside the country). So please do join in, I am planning to put a lot of stuff there, tutorials, highres versions of all my photos, raw files etc. Also check out my gallery.
https://www.patreon.com/c/AurelManea
Coming back to the subject : For this particular method to work you need a good video card with plenty of vram. My video card can upscale from 2k to 5k in around 15 minutes (an nvidia rtx 3090 with 24GB ram). The more ram you have the higher resolution you can get. I use this for crops from photos so don’t feel I ruined the resolution by cropping or guilt that I do not have a lens of more than 420mm equivalent (I have Olympus gear). It is unlikely that you will have enough memory to go from 5k to 10k using this or the time. This is something to recover from crops to full resolution and it is meant for very important shots to get the best quality. For generic fast upscales but medicre quality there is Topaz.
Link for comfyUI here : https://github.com/comfyanonymous/ComfyUI
Youtube is full of tutorials about anything in it from image generation, video, upscaling etc.
Here are some more examples of how I used this in actual photos
