Gemini Flash 2.0 Image generation

In this video, I explore Google's new Gemini Flash 2.0 Experimental model - an AI capable of generating both images and text simultaneously. I test various prompts to create minimalist designs.

I experiment with different animal objects (snail, elephant, dog, cat), paper craft aesthetics, font styles, and even complex scenes featuring a moonlit pond and a semi-robotic dragon with wires and connections.

As I iterate through multiple designs, I showcase both the strengths and limitations of the model - highlighting how image quality sometimes degrades with continuous editing, how the AI occasionally misinterprets prompts, and how starting fresh often yields better results than continued modifications.

In the end, I come up with an image I really like and will use in my creative work.

Voice to text I use:

Wispr flow → wisprflow.ai/r/GLEB3 — get $15 discount
LinkedIn →   / glebkalinin  
Tool Building Ape (Russian) → https://t.me/toolusing/