So, I decided to mess around with the new Gemini API's image editing feature, which is still in beta (experimental). Thought I'd give it a shot using Lovable AI—and honestly, I was surprised. I didn’t touch a single line of code, just typed out 7–8 prompts, and boom—the whole thing was done in under 20 minutes.
Lovable AI handled everything—UI design, functionality, and even the little details. I described what I wanted, and it just built it. The layout’s clean, the features are solid, and it doesn’t feel like one of those rushed AI builds.
Now about the app itself:
- You type something like "Add a Mickey Mouse frame on the wall"—and it does exactly that.
- Or say "Make him wear black goggles that fit perfectly."—and it just works.
- The Gemini API is surprisingly accurate with visual edits, even for detailed stuff like positioning or adding accessories.
I didn’t expect it to turn out this smooth, especially since this API is still experimental. But Lovable AI handled it like a champ. If you’re curious, check it out here: Glassy Image Alchemist
