Good start but...
Let's start with the negatives:
1. Stable Diffusion: It seems you are using it... not the best model and the results are subpar. Either improve it or provide a choice of other models; I don’t know, but what we have now looks quite unsatisfactory.
2. Request Processing Time: Each subsequent request and edit in the thread takes longer than the previous one. After five edits, you're waiting about two minutes and receiving worse results than before.
3. Versioning of Images: I want to go back to the version that was three steps ago and start making changes from there, removing some edits (a very useful feature that is currently lacking). Right now, the neural network easily messes up works, forcing you to start over.
Now for the positives:
1. User-friendly Interface: It's convenient to use.
2. Quick and Easy Start: It’s genuinely fast and simple to begin.
3. Seems Free: I couldn't find any information about models and limits, which suggests it's free. It appears to be running on my device, not on a server, as I see no confirmation otherwise and I'm unsure how it works.
In general, it's an interesting project. Keep developing! ❤️