I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
enlarger.appFull disclaimer: the Enlarger app is something I did.
Fuller disclaimer: this kind of a "postmark"-sized pic blowup test case is borderline unfair to every model in the comparison, mine included.
You could have also disclosed the fact that the model that you call "Enlarger — Base" is a free, open-weights model named 4xNomosWebPhoto, given that this post is apparently comparing the models. You seem to even explicitly claim it as your model in this comment.
Thanks for the feedback, I explicitly added that now to the blog post. The wording on this comment was off. That model is listed in the "Acknowledgements"-tab of the application.
The end goal on working on this program has been that the user shouldn't care much about what model/architecture/whatever tech is used, and should be abstracted away as much as possible. Though in a blog post like this it should be mentioned, it was hard to break away from that self-imposed mental jail :)
Enlarger's ear lobes are molting!
Most any upsize looks better if you toss hue protected noise back in; you've made good use of that point.
Thanks. Yeah, I used to do the tug-and-pull of scaling, then adding noise and sharpening, scaling a little more, adding more noise, adding more sharpening, etc. to ad nauseam. And photoshop sharpening was too much always for my taste.
At scale, this tool should speed that process up a "bit". Let's say 100s of photos.
I can only see a mouth and nose on mobile
Check the 2nd note of the blog post (best viewed with desktop)
A human face is the easiest target - because AI upscalers can make up a "fake" face realistic enough to fool most people.
Especially something as standardized as a passport photo.
See this article shared on HN a week ago:
https://blog.jimgrey.net/2026/06/30/what-happens-when-the-in...
I disagree - the human face is the hardest target exactly because (most) AI upscalers make up face traits and alter expressions.
The blog post you linked uses ChatGPT, which uses generative cloud AI with billions of parameters, which does indeed make up stuff.
The models that are diff'd on this blog post uses a whole different local architecture which has nothing to do with ChatGPT or Nano Banana 2, which the likes the blog post critiques. It is more rooted in older tech pre-generative era.
I do agree with the blog post, and "fool most people" -effect is especially strong when it's an old photo from a time that is not freshly in the viewers memory, and the subject no longer looks like it.
On mobile it shows a zoomed in version of the image. It shows the small details even more. I was not really agreeing with the author on the superiority of Enlarger.
With the full images, the comparison is much more to the advantage of Enlarger.
Is it me, or do all upscaled photos look like it's a video game, or the face is made out of smooth plastic?
Pet peeve: people who use "4x" instead of "quadrupled".
It is what it is with the title length