I started exploring the NFTs, Blockchain & AI / ML with Deep Learning in a little more depth.
That’s when I found out about the VQGAN + CLIP, CLIP + Guided Diffusion + VGQAN, etc.

I mean it is so amazing!
Like you input relevant text to the model & it creates the relevant stuff (depending on the image data set used to train).
AFAIK, the dataset used here was CC12M (I may be wrong).

So I tried my hands on a few different moded codebases on Google Colab.
After a lot of trial error, experiments, I got relevant results which I obviously minted on OpenSea.

Some examples are :